Table of Contents

1. Introduction
2. Hardware requirements and dependencies
3. FusionCatcher in scientific articles
4. Installation and usage examples
5. Quick installation
6. Usage
7. Aligners
8. Command line options
9. Methods
10. Comparisons to other tools
11. License
12. Citing
13. Reporting bugs

1 - INTRODUCTION

FusionCatcher searchers for somatic novel/known fusion genes, translocations and/or chimeras in RNA-seq data (stranded/unstranded paired-end/single-end reads FASTQ files produced by Illumina next-generation sequencing platforms like Illumina Solexa/HiSeq/NextSeq/MiSeq/MiniSeq) from diseased samples.

The aims of FusionCatcher are: * very good detection rate for finding candidate somatic fusion genes (see somatic mutations; using a matched normal sample is optional; several databases of known fusion genes found in healthy samples are used as a list of known false positives; biological knowledge is used, like for example gene fusion between a gene and its pseudogene is filtered out), * very good RT-PCR validation rate of found candidate somatic fusion genes (this is very important for us), * very easy to use (i.e. no a priori knowledge of bioinformatic databases and bioinformatics is needed in order to run FusionCatcher BUT Linux/Unix knowledge is needed; it allows a very high level of control for expert users), * very good detection of challenging fusion genes, like for example IGH fusions, CIC fusions, DUX4 fusions, CRLF2 fusions, TCF3 fusions, etc., * to be as automatic as possible (i.e. the FusionCatcher will choose automatically the best parameters in order to find candidate somatic fusion genes, e.g. finding automatically the adapters, quality trimming of reads, building the exon-exon junctions automatically based on the length of the reads given as input, etc. while giving also full control to expert users) while providing the best possible detection rate for finding somatic fusion genes (with a very low rate of false positives but a very good sensitivity).

FusionCatcher supports: * as input FASTQ and/or SRA file types (paired-end reads from stranded or strand-specific experiments, single-end reads when they are longer than 130bp), * five different methods (using Bowtie aligner and optionally BLAT, STAR, BOWTIE2 aligners) for finding new fusion genes BUT by default only Bowtie, Blat, and STAR aligners will be used, * several eukaryotic organisms (which are in Ensembl database), like for example, human, rat, mouse, dog, etc.

2 - HARDWARE REQUIREMENTS AND DEPENDENCIES

For running FusionCatcher it is needed a computer with: * 64-bit *NIX environment * minimum 24 GB of RAM (in many cases it might work even with 16GB of RAM for very small input FASTQ files in order of megabytes) * 1 CPU (minimum) * ~700 GB temporary disk space (needed just for temporary files)

2.1 - Required dependencies

• Linux/Unix 64-bit (e.g. Ubuntu version 12.04/14.04/16.04 or newer)
• Python version 2.7.x (>=2.6.0 and < 3.0 is fine)
• BioPython version 1.73 (>=1.50 is fine)
• Bowtie 64-bit version 1.2.3 http://bowtie-bio.sourceforge.net/index.shtml (will be installed by boostrap.py)
• forked SeqTK version 1.2-r101c-dirty http://github.com/ndaniel/seqtk (will be installed by boostrap.py)
• organism specific data from Ensembl database release 98 (all downloading and the necessary building process is handled automatically by the included/provided tool fusioncatcher-build and therefore no knowledge of Ensembl database or other databases is needed) (will be installed by boostrap.py)
• STAR version 2.7.2b https://github.com/alexdobin/STAR . Executables are available at http://github.com/alexdobin/STAR/releases (will be installed by boostrap.py)
• BOWTIE2 version 2.3.5.1 http://bowtie-bio.sourceforge.net/bowtie2/index.shtml (will be installed by boostrap.py)
• BBMAP version 38.44 https://sourceforge.net/projects/bbmap/ (will be installed by boostrap.py)

2.2 - Optional dependencies

2.2.1 - Strongly recommended

These are expected by default to be installed but their use can be disabled by using the command line option ‘–skip-blat’. * BLAT version 0.35 http://users.soe.ucsc.edu/~kent/src/ . Executables are available at http://hgdownload.cse.ucsc.edu/admin/exe/ . Please, check the license to see if it allows you to run/use it! This is needed by FusionCatcher (hint: if you are a non-profit organization you should be fine) (will be installed by boostrap.py) * faToTwoBit http://users.soe.ucsc.edu/~kent/src/ . Executables are available at http://hgdownload.cse.ucsc.edu/admin/exe/ . Please, check the license to see if it allows you to run/use it! This is needed by FusionCatcher and fusioncatcher-build if one plans to use BLAT as a second (optional) alternative method for finding fusion genes! (required also by option --blat-visualization) (will be installed by boostrap.py)

Note: If one does not want to install BLAT (whilst installing FusionCatcher automatically thru bootstrap.py) and also not to use BLAT with FusionCatcher then using command line -k option of bootstrap.py will do that.

2.2.2 - Nice to have (but optional)

• Velvet (de novo assembler) version 1.2.10 http://www.ebi.ac.uk/~zerbino/velvet/ . This is needed if one plans to do de novo assembly of the reads which support the candidate fusion genes. (required by option --assembly of FusionCatcher) (will be installed by boostrap.py)
• fastq-dump version 2.9.6 from NCBI SRA Toolkit http://www.ncbi.nlm.nih.gov/Traces/sra/?view=software . This is needed by FusionCatcher if one plans to use as input SRA files (will be installed by boostrap.py)
• Python library openpyxl version 1.5.6 http://pypi.python.org/pypi/openpyxl (other versions might work but have not been tested). It is needed by fusioncatcher-build for parsing the ConjoinG database.
• Python library xlrd version 0.6.1 http://pypi.python.org/pypi/xlrd (other versions might work but have not been tested). It is needed by fusioncatcher-build for parsing the ChimerDB database.
• pigz version 2.3.1 http://zlib.net/pigz/ for using GZIP on several CPUs in parallel (other older versions might support this) (will be installed by boostrap.py)
• Picard tools version 2.21.2 http://github.com/broadinstitute/picard (will be installed by boostrap.py)

2.3 - Genomic Databases

These are used (downloaded and parsed) automatically by boostrap.py of FusionCatcher: * ENSEMBL database http://www.ensembl.org/ (required) * UCSC database http://hgdownload.cse.ucsc.edu/downloads.html#human (required) * RefSeq database (thru UCSC database) (required) * Viruses/bacteria/phages genomes database (from the NCBI database) ftp://ftp.ncbi.nlm.nih.gov/genomes/Viruses/ (required) * COSMIC database http://cancer.sanger.ac.uk/cancergenome/projects/cosmic/ (optional) * TICdb database http://www.unav.es/genetica/TICdb/ (optional) * ChimerDB 2.0 database (literature-based annotation) http://ercsb.ewha.ac.kr/FusionGene/ (optional) * Cancer Genome Project (CGP) translocations database http://www.sanger.ac.uk/genetics/CGP/Census/ (optional) * ConjoinG database http://metasystems.riken.jp/conjoing/ (optional) * CACG conjoined genes database http://cgc.kribb.re.kr/map/ (optional) * DGD database http://dgd.genouest.org/ (optional) * Illumina BodyMap2 RNA-seq database http://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-513/

NOTES: * ENSEMBL database is used for finding novel/known fusions genes * COSMIC, TICdb, ChimerDB, Cancer Genome Project, ConjoinG, and manual curated fusion gene database are indexed and used further for annotating/labeling the found fusion genes for an easier visualization of novel genes (i.e. not published yet) found by FusionCatcher. For more information how this is used see Tables 1,2,3. * FusionCatcher can work just fine and is able to find fusion genes without any of the optional dependencies/tools/programs! * if BLAT is not installed (or one does not want to use it) please use option ‘–skip-blat’ in order to let know FusionCatcher that it should not use it! Also, specifying that BLAT should be be used can be done by editing manually the line aligners = blat,star from file fusioncatcher/etc/configuration.cfg (that is remove the blat string).

3 - FusionCatcher in scientific articles

FusionCatcher has been used for finding novel and known fusion genes in the following articles: * S. Kangaspeska, S. Hultsch, H. Edgren, D. Nicorici, A. Murumägi, O.P. Kallioniemi, Reanalysis of RNA-sequencing data reveals several additional fusion genes with multiple isoforms, PLOS One, Oct. 2012. http://dx.doi.org/10.1371/journal.pone.0048745 * H. Edgren, A. Murumagi, S. Kangaspeska, D. Nicorici, V. Hongisto, K. Kleivi, I.H. Rye, S. Nyberg, M. Wolf, A.L. Borresen-Dale, O.P. Kallioniemi, Identification of fusion genes in breast cancer by paired-end RNA-sequencing, Genome Biology, Vol. 12, Jan. 2011. http://dx.doi.org/10.1186/gb-2011-12-1-r6 * JN. Honeyman, EP. Simon, N. Robine, R. Chiaroni-Clarke, DG. Darcy, I. Isabel, P. Lim, CE. Gleason, JM. Murphy, BR. Rosenberg, L. Teegan, CN. Takacs, S. Botero, R. Belote, S. Germer, A-K. Emde, V. Vacic, U. Bhanot, MP. LaQuaglia, and S.M. Simon, Detection of a Recurrent DNAJB1-PRKACA Chimeric Transcript in Fibrolamellar Hepatocellular Carcinoma, Science 343 (6174), Feb. 2014, pp. 1010-1014, http://dx.doi.org/10.1126/science.1249484 * T. Pietsch, I. Wohlers, T. Goschzik, V. Dreschmann, D. Denkhaus, E. Dorner, S. Rahmann, L. Klein-Hitpass, Supratentorial ependymomas of childhood carry C11orf95–RELA fusions leading to pathological activation of the NF-kB signaling pathway, Acta Neuropathologica 127(4), Apr. 2014, pp. 609-611. http://dx.doi.org/10.1007/s00401-014-1264-4 * M. Jimbo, K.E. Knudsen, J.R. Brody, Fusing Transcriptomics to Progressive Prostate Cancer, The American Journal of Pathology, 2014, http://dx.doi.org/10.1016/j.ajpath.2014.08.001 * Y.P. Yu, Y. Ding, Z. Chen, S. Liu, A. Michalopoulos, R. Chen, Z. Gulzar, B. Yang, K.M. Cieply, A. Luvison, B.G. Ren, J.D. Brooks, D. Jarrard, J.B. Nelson. G.K. Michalopoulos, G.C. Tseng, J.H. Luo, Novel fusion transcripts associate with progressive prostate cancer, The American Journal of Pathology, 2014, http://dx.doi.org/10.1016/j.ajpath.2014.06.025 * C.M Lindqvist, J. Nordlund, D. Ekman, A. Johansson, B.T. Moghadam, A. Raine, E. Overnas, J. Dahlberg, P. Wahlberg, N. Henriksson, J. Abrahamsson, B.M. Frost, D. Grander, M. Heyman, Rolf Larsson, J. Palle, S. Soderhall, E. Forestier, G. Lonnerholm, A.C. Syvanen, E.C. Berglund, The Mutational Landscape in Pediatric Acute Lymphoblastic Leukemia Deciphered by Whole Genome Sequencing, Human Mutation, 2014, http://dx.doi.org/10.1002/humu.22719 * I. Panagopoulos, L. Gorunova, B. Davidson, Sverre Heim, Novel TNS3-MAP3K3 and ZFPM2-ELF5 fusion genes identified by RNA sequencing in multicystic mesothelioma with t(7;17)(p12;q23) and t(8;11)(q23;p13), Cancer Letters, Dec. 2014, http://dx.doi.org/10.1016/j.canlet.2014.12.002 * J.C Lee, Y.M. Jeng, S.Y. Su, C.T Wu, K.S. Tsai, C.H. Lee, C.Y. Lin, J.M. Carter, J. W. Huang, S.H. Chen, S.R. Shih, A. Marino-Enriquez, C.C. Chen, A.L. Folpe, Y.L. Chang, C.W. Liang, Identification of a novel FN1–FGFR1 genetic fusion as a frequent event in phosphaturic mesenchymal tumour, The journal of Pathology, Jan. 2015, http://dx.doi.org/10.1002/path.4465 * J. Nordlund, C.L. Backlin, V. Zachariadis, L. Cavelier, J. Dahlberg, I. Ofverholm, G. Barbany, A. Nordgren, E. Overnas, J. Abrahamsson, T. Flaegstad, M.M. Heyman, O.G. Jonsson, J. Kanerva, R. Larsson, J. Palle, K. Schmiegelow, M.G. Gustafsson, G. Lonnerholm, E. Forestier, A.C. Syvanen, DNA methylation-based subtype prediction for pediatric acute lymphoblastic leukemia, Clinical Epigenetics, 7:11, Feb. 2015, http://dx.doi.org/10.1186/s13148-014-0039-z * J.H. Luo, S. Liu, Z.H. Zuo, R. Chen, G.C. Tseng, Y.P. Yu, Discovery and Classification of Fusion Transcripts in Prostate Cancer and Normal Prostate Tissue, The American Journal of Pathology, May 2015, http://dx.doi.org/10.1016/j.ajpath.2015.03.008 * T. Meissner, K.M. Fisch, L. Gioia, OncoRep: an n-of-1 reporting tool to support genome-guided treatment for breast cancer patients using RNA-sequencing, BMC Medical Genomics, May 2015, http://dx.doi.org/10.1186/s12920-015-0095-z * S. Torkildsen, L. Gorunova, K. Beiske, G.E. Tjonnfjord, S. Heim, I. Panagopoulos, Novel ZEB2-BCL11B Fusion Gene Identified by RNA-Sequencing in Acute Myeloid Leukemia with t(2;14)(q22;q32), PLOS One, July 2015, http://dx.doi.org/10.1371/journal.pone.0132736 * M. Cieslik, R. Chugh, Y.M. Wu, M. Wu, C. Brennan, R. Lonigro, F. Su, R. Wang, J. Siddiqui, R. Mehra, X. Cao, D. Lucas, A.M. Chinnaiyan, D. Robinson, The use of exome capture RNA-seq for highly degraded RNA with application to clinical cancer sequencing, Genome Research, August 2015, http://dx.doi.org/10.1101/gr.189621.115 * E.P. Simon, C.A. Freije, B.A. Farber, G. Lalazar, D.G. Darcy, J.N. Honeyman, R. Chiaroni-Clark, B.D. Dill, H. Molina, U.K. Bhanot, M.P. La Quaglia, B.R. Rosenberg, S.M. Simon, Transcriptomic characterization of fibrolamellar hepatocellular carcinoma, PNAS, October 2015, http://dx.doi.org/10.1073/pnas.1424894112 * Y. Marincevic-Zuniga, V. Zachariadis, L. Cavelier, A. Castor, G. Barbany, E. Forestier, L. Fogelstrand, M. Heyman, J. Abrahamsson, G. Lonnerholm, A. Nordgren, A.C. Syvanen, J. Nordlund, PAX5-ESRRB is a recurrent fusion gene in B-cell precursor pediatric acute lymphoblastic leukemia, Haematologica, October 2015, http://dx.doi.org/10.3324/haematol.2015.132332 * M. Brenca, S. Rossi, M. Polano, D. Gasparotto, L. Zanatta, D. Racanelli, L. Valori, S. Lamon, A.P. Dei Tos, R. Maestro, Transcriptome sequencing identifies ETV6-NTRK3 as a gene fusion involved in GIST, The Journal of Pathology, November 2015, http://dx.doi.org/10.1002/path.4677 * Roberts K.G., et al. High Frequency and Poor Outcome of Ph-like Acute Lymphoblastic Leukemia in Adults, Blood Journal, Vol. 126, December 2015, http://www.bloodjournal.org/content/126/23/2618 * Kekeeva T., et al. Novel fusion transcripts in bladder cancer identified by RNA-seq, Cancer Letters, Feb. 2016, http://dx.doi.org/10.1016/j.canlet.2016.02.010 * Panagopoulos I. et al. Rare MLL-ELL fusion transcripts in childhood acute myeloid leukemia-association with young age and myeloid sarcomas?, Experimental Hematology & Oncology, Vol. 5, March 2016, http://dx.doi.org/10.1186/s40164-016-0037-2 * Chang W. et al, Multi-Dimensional ClinOmics for Precision Therapy of Children and Adolescent Young Adults with Relapsed and Refractory Cancer: A report from the Center for Cancer Research, Clinical Cancer Research, March 2016, http://dx.doi.org/10.1158/1078-0432.CCR-15-2717 * Spans L. et al., Recurrent MALAT1-GLI1 oncogenic fusion and GLI1 upregulation define a subset of plexiform fibromyxoma, The Journal of Pathology, April 2016, http://dx.doi.org/10.1002/path.4730 * Micci F. et al., Cytogenetic and Molecular Profile of Endometrial Stromal Sarcoma, Genes Chromosomes & Cancer, May 2016, http://dx.doi.org/10.1002/gcc.22380 * Olsen K.T. et al., Novel Fusion Genes and Chimeric Transcripts in Ependymal Tumors, Genes Chromosomes & Cancer, July 2016, http://dx.doi.org/10.1002/gcc.22392 * Panagopoulos I. et al., Recurrent fusion of the genes FN1 and ALK in gastrointestinal leiomyomas, Modern Pathology, July 2016, http://dx.doi.org/10.1038/modpathol.2016.129 * Barnes D.J. et al., A germline mutation of CDKN2A and a novel RPLP1-C19MC fusion detected in a rare melanotic neuroectodermal tumor of infancy: a case report, BMC Cancer, August 2016, http://dx.doi.org/10.1186/s12885-016-2669-3 * Panagopoulos I. et al., Gene fusions AHRR-NCOA2, NCOA2-ETV4, ETV4-AHRR, P4HA2-TBCK, and TBCK-P4HA2 resulting from the translocations t(5;8;17)(p15;q13;q21) and t(4;5)(q24;q31) in a soft tissue angiofibroma, Oncology Reports, Sept. 2016, http://dx.doi.org/10.3892/or.2016.5096 * Lang P.Y. et al., ATR maintains chromosomal integrity during postnatal cerebellar neurogenesis and is required for medulloblastoma formation, Development, Nov. 2016, http://dx.doi.org/10.1242/dev.139022 * Gu Z. et al., Genomic analyses identify recurrent MEF2D fusions in acute lymphoblastic leukaemia, Nature Communications, Nov. 2016, http://dx.doi.org/10.1038/ncomms13331 * Alaei-Mahabadi B. et al., Global analysis of somatic structural genomic alterations and their impact on gene expression in diverse human cancers, PNAS, Nov. 2016, http://dx.doi.org/10.1073/pnas.1606220113 * Yap K.L. et al., Diagnostic evaluation of RNA sequencing for the detection of genetic abnormalities associated with Ph-like acute lymphoblastic leukemia (ALL), Leukemia & Lymphoma, Nov. 2016, http://dx.doi.org/10.1080/10428194.2016.1219902 * Brunetti M. et al., Recurrent fusion transcripts in squamous cell carcinomas of the vulva, Oncotarget, Feb. 2017, http://dx.doi.org/10.18632/oncotarget.15167 * Sahm F. et al., DNA methylation-based classification and grading system for meningioma: a multicentre, retrospective analysis, The Lancet Oncology, March 2017, http://dx.doi.org/10.1016/S1470-2045(17)30155-9 * Reshmi S.C. et al., Targetable kinase gene fusions in high risk B-ALL: a study from the Children’s Oncology Group, Blood, April 2017, https://doi.org/10.1182/blood-2016-12-758979 * Tomic T.T. et al., A new GTF2I-BRAF fusion mediating MAPK pathway activation in pilocytic astrocytoma, PLOS One, April 2017, https://doi.org/10.1371/journal.pone.0175638 * Schwartzman O. et al., Suppressors and activators of JAK-STAT signaling at diagnosis and relapse of acute lymphoblastic leukemia in Down syndrome, PNAS, May 2017, https://doi.org/10.1073/pnas.1702489114 * Panagopoulos I. et al., Identification of SETD2-NF1 fusion gene in a pediatric spindle cell tumor with the chromosomal translocation t(3;17)(p21;q12), Oncology Reports, May 2017, https://doi.org/10.3892/or.2017.5628 * Sorenson E.C. et al., Genome and transcriptome profiling of fibrolamellar hepatocellular carcinoma demonstrates p53 and IGF2BP1 dysregulation, PLOS One, May 2017, https://doi.org/10.1371/journal.pone.0176562 * Hettmer S. et al., Epithelioid hemangioendotheliomas of the liver and lung in children and adolescents, Pediatric Blood & Cancer, June 2017, https://doi.org/10.1002/pbc.26675 * Pisapia D.J. et al., Next-Generation Rapid Autopsies Enable Tumor Evolution Tracking and Generation of Preclinical Models, JCO Precision Oncology, June 2017, https://doi.org/10.1200/PO.16.00038 * Bekers E.M. et al., Soft tissue angiofibroma – clinicopathologic, immunohistochemical and molecular analysis of 14 cases, Genes, Chromosomes and Cancers, June 1027, https://doi.org/10.1002/gcc.22478 * Mandahl N. et al., Scattered genomic amplification in dedifferentiated liposarcoma, Molecular Cytogenetics, Vol. 10, June 2017, https://doi.org/10.1038/10.1186/s13039-017-0325-5 * Biaconi D. et al., Biochemical and genetic predictors of overall survival in patients with metastatic pancreatic cancer treated with capecitabine and nab-paclitaxel, Scientific Reports, Vol. 7, July 2017, https://doi.org/10.1038/s41598-017-04743-0 * Sahraeian S.M.E. et.al., Gaining comprehensive biological insight into the transcriptome by performing a broad-spectrum RNA-seq analysis, Nature Communication, Vol. 8, July 2017, https://doi.org/10.1038/s41467-017-00050-4 * Micci F. et al., Fusion of the Genes BRD8 and PHF1 in Endometrial Stromal Sarcoma, Genes, Chromosomes and Cancers, July 2017, https://doi.org/10.1002/gcc.22485 * Kivioja J.L. et al., Chimeric NUP98–NSD1 transcripts from the cryptic t(5;11)(q35.2;p15.4) in adult de novo acute myeloid leukemia, Leukemia and Lymphoma, Aug. 2017, http://dx.doi.org/10.1080/10428194.2017.1357174 * Marincevic-Zuniga Y. et al., Transcriptome sequencing in pediatric acute lymphoblastic leukemia identifies fusion genes associated with distinct DNA methylation profiles, Journal of Hematology and Oncology, Aug. 2017, http://doi.org/10.1186/s13045-017-0515-y * Kumar A. et al., The impact of RNA sequence library construction protocols on transcriptomic profiling of leukemia, BMC Genomics, Aug. 2017, http://doi.org/10.1186/s12864-017-4039-1 * Singh R. et al., Analysis of the whole transcriptome from gingivo-buccal squamous cell carcinoma reveals deregulated immune landscape and suggests targets for immunotherapy, PLOS One, Sept. 2017, http://doi.org/10.1371/journal.pone.0183606 * Arbajian E. et al., Inflammatory leiomyosarcoma is a distinct tumor characterized by near-haploidization, few somatic mutations, and a primitive myogenic gene expression signature, Modern Pathology, Sept. 2017, http://doi.org/10.1038/modpathol.2017.113 * Varghese, A.M. et al., Clinical and molecular characterization of patients with cancers of unknown primary in the modern era, Annals of Oncology, Sep. 2017, http://doi.org/10.1093/annonc/mdx545 * Persson H. et al., Frequent miRNA-convergent fusion gene events in breast cancer, Nature Communications, Oct. 2017, http://doi.org/doi:10.1038/s41467-017-01176-1 * Gu J. et al., RNA-seq Based Transcription Characterization of Fusion Breakpoints as a Potential Estimator for Its Oncogenic Potential, BioMed Research International, Oct. 2017, http://doi.org/10.1155/2017/9829175 * Dalin M.G. et al., Multi-dimensional genomic analysis of myoepithelial carcinoma identifies prevalent oncogenic gene fusions, Nature Communications, Oct. 2017, http://doi.org/10.1038/s41467-017-01178-z * Panagopoulos I. et al., DEK-NUP214-Fusion Identified by RNA-Sequencing of an Acute Myeloid Leukemia with t(9;12)(q34;q15), Cancer Genomics and Proteomics, Nov. 2017, http://doi.org/10.21873/cgp.20053 * Kita K. et al., In vivo imaging xenograft models for the evaluation of anti-brain tumor efficacy of targeted drugs, Cancer Medicine, Nov. 2017, http://doi.org/10.1002/cam4.1255 * Xia Q. Y. et al., Novel gene fusion of PRCC-MITF defines a new member of MiT family translocation renal cell carcinoma: clinicopathologic analysis and detection of the gene fusion by RNA-sequencing and FISH, Histopathology, Nov. 2017, http://doi.org/10.1111/his.13439 * Chang M.T. et al., Small cell carcinomas of the bladder and lung are characterized by a convergent but distinct pathogenesis, Clinical Cancer Research, Nov. 2017, http://doi.org/10.1158/1078-0432.CCR-17-2655 * Arbajian E. et al., In-depth Genetic Analysis of Sclerosing Epithelioid Fibrosarcoma Reveals Recurrent Genomic Alterations and Potential Treatment Targets, Clinical Cancer Research, Dec. 2017, http://doi.org/10.1158/1078-0432.CCR-17-1856 * Brunetti M. et al., RNA-Sequencing identifies novel GREB1-NCOA2 fusion gene in a uterine sarcoma with the chromosomal translocation t(2;8)(p25;q13), Genes Chromosomes Cancer, Dec. 2017, http://doi.org/10.1002/gcc.22518 * Gillard M. et al., Integrative Genomic Analysis of Coincident Cancer Foci Implicates CTNNB1 and PTEN Alterations in Ductal Prostate Cancer, European Urology Focus, Dec. 2017, https://doi.org/10.1016/j.euf.2017.12.003 * Alamri A.M. et al., Expanding primary cells from mucoepidermoid and other salivary gland neoplasms for genetic and chemosensitivity testing, Disease Models and Mechanisms, Jan. 2018, https://doi.org/10.1242/dmm.031716 * Togashi Y. et al., MYB and MYBL1 in adenoid cystic carcinoma: diversity in the mode of genomic rearrangement and transcripts, Modern Pathology, Feb. 2018, https://doi.org/10.1038/s41379-018-0008-8 * Roussy M. et al., NUP98-BPTF gene fusion identified in primary refractory acute megakaryoblastic leukemia of infancy, Genes, Chromosomes and Cancer, Feb. 2018, https://doi.org/10.1002/gcc.22532 * Nishiyama A. et al., Foretinib overcomes entrectinib resistance associated with the NTRK1 G667C mutation in NTRK1 fusion-positive tumor cells in a brain metastasis model, Clinical Cancer Research, Feb. 2018, https://doi.org/10.1158/1078-0432.CCR-17-1623 * Liu H. et al., Identifying and Targeting Sporadic Oncogenic Genetic Aberrations in Mouse Models of Triple-Negative Breast Cancer, Cancer Discovery, March 2018, https://doi.org/10.1158/2159-8290.CD-17-0679 * Panagopoulos I. et al., PAN3–PSMA2 fusion resulting from a novel t(7;13)(p14;q12) chromosome translocation in a myelodysplastic syndrome that evolved into acute myeloid leukemia, Experimental Hematology and Oncology, March 2018, https://doi.org/10.1186/s40164-018-0099-4 * Alholle A. et al., Genetic analyses of undifferentiated small round cell sarcoma identifies a novel sarcoma subtype with a recurrent CRTC1‐SS18 gene fusion, The Journal of Pathology, March 2018, https://doi.org/10.1002/path.5071 * Agostini A. et al., Identification of novel cyclin gene fusion transcripts in endometrioid ovarian carcinomas, Internation Journal of Cancer, April 2018, https://doi.org/10.1002/ijc.31418 * Brunetti M. et al., Identification of an EPC2-PHF1 fusion transcript in low-grade endometrial stromal sarcoma, Oncotarget, April 2018, https://doi.org/10.18632/oncotarget.24969 * Panagopoulos I. et al., RUNX1-PDCD6 fusion resulting from a novel t(5;21)(p15;q22) chromosome translocation in myelodysplastic syndrome secondary to chronic lymphocytic leukemia, PLOS One, April 2018, https://doi.org/10.1371/journal.pone.0196181 * Bie D.J. et al., Single-cell sequencing reveals the origin and the order of mutation acquisition in T-cell acute lymphoblastic leukemia, Leukemia, April 2018, https://doi.org/10.1038/s41375-018-0127-8 * Chiaretti S. et al., Rapid identification of BCR/ABL1‐like acute lymphoblastic leukaemia patients using a predictive statistical model based on quantitative real time‐polymerase chain reaction: clinical, prognostic and therapeutic implications, British Journal of Haematology, April 2018, https://doi.org/10.1111/bjh.15251 * Wang X.T. et al., RNA sequencing of Xp11 translocation-associated cancers reveals novel gene fusions and distinctive clinicopathologic correlations, Modern Pathology, April 2018, https://doi.org/10.1038/s41379-018-0051-5 * Churchman M.L. et al., Germline Genetic IKZF1 Variation and Predisposition to Childhood Acute Lymphoblastic Leukemia, Cancer Cell, May 2018, https://doi.org/10.1016/j.ccell.2018.03.021 * Knuuttila M. et al., Intratumoral androgen levels are linked to TMPRSS2-ERG fusion in prostate cancer, Endocrinology-related-Related Cancer, May 2018, doi: https://doi.org/10.1530/ERC-18-0148 * He H., et al., Identification of a Recurrent LMO7–BRAF Fusion in Papillary Thyroid Carcinoma, Thyroid, May 2018, https://doi.org/10.1089/thy.2017.0258 * Wang C. et al., Whole-genome sequencing reveals genomic signatures associated with the inflammatory microenvironments in Chinese NSCLC patients, Nature Communications, May 2018, https://doi.org/10.1038/s41467-018-04492-2 * Parris T.Z. et al., Genome-wide multi-omics profiling of the 8p11-p12 amplicon in breast carcinoma, Oncotarget, May 2018, https://dx.doi.org/10.18632%2Foncotarget.25329 * Gioiosa S. et al., Massive NGS Data Analysis Reveals Hundreds Of Potential Novel Gene Fusions in Human Cell Lines, GigaScience, June 2018, https://doi.org/10.1093/gigascience/giy062 * Roberts, K.G. et al., Genomic and outcome analyses of Ph-like ALL in NCI standard-risk patients: a report from the Children’s Oncology Group, Blood, June 2018, https://doi.org/10.1182/blood-2018-04-841676 * Rosenberg S. et al., A recurrent point mutation in PRKCA is a hallmark of chordoid gliomas, Nature Communications, June 2018, https://doi.org/10.1038/s41467-018-04622-w * Ganly I. et al., Integrated Genomic Analysis of Hürthle Cell Cancer Reveals Oncogenic Drivers, Recurrent Mitochondrial Mutations, and Unique Chromosomal Landscapes, Cancer Cell, August 2018, https://doi.org/10.1016/j.ccell.2018.07.002 * Wrzeszczynski K.O. et al., Analytical Validation of Clinical Whole-Genome and Transcriptome Sequencing of Patient Derived Tumors Clinical Application of Whole-Genome Sequencing for Reporting Targetable Variants in Cancer, The Journal of Molecular Diagnostics, Aug. 2018, https://doi.org/10.1016/j.jmoldx.2018.06.007 * Hultsch S. et al., Association of tamoxifen resistance and lipid reprogramming in breast cancer, BMC Cancer, Aug. 2018, https://doi.org/10.1186/s12885-018-4757-z * Alexander T.B. et al., The genetic basis and cell of origin of mixed phenotype acute leukaemia, Nature, Sept. 2018, https://doi.org/10.1038/s41586-018-0436-0 * Al-Ibraheemi A. et al., Aberrant receptor tyrosine kinase signaling in lipofibromatosis: a clinicopathological and molecular genetic study of 20 cases, Modern Pathology, Oct. 2018, https://doi.org/10.1038/s41379-018-0150-3 * Engqvist H. et al., Transcriptomic and genomic profiling of early-stage ovarian carcinomas associated with histotype and overall survival, Oncotarget, Oct. 2018, https://doi.org/10.18632/oncotarget.26225 * Haider Z. et al., An integrated transcriptome analysis in T‐cell acute lymphoblastic leukemia links DNA methylation subgroups to dysregulated TAL1 and ANTP homeobox gene expression, Cancer Medicine, Dec. 2018, https://doi.org/10.1002/cam4.1917 * Sekimizu M. et al., Frequent mutations of genes encoding vacuolar H+‐ATPase components in granular cell tumors, Genes Chromosomes & Cancer, Dec. 2018, https://doi.org/10.1002/gcc.22727 * Troll C.J. et al., Structural Variation Detection by Proximity Ligation from Formalin-Fixed, Paraffin-Embedded Tumor Tissue, The journal of molecular diagnostic, Dec. 2018, https://doi.org/10.1016/j.jmoldx.2018.11.003 * Xiao W. et al., PHF6 and DNMT3A mutations are enriched in distinct subgroups of mixed phenotype acute leukemia with T-lineage differentiation, Blood Advances, Dec. 2018, https://doi.org/10.1182/bloodadvances.2018023531 * Diolaiti D. et al., A recurrent novel MGA–NUTM1 fusion identifies a new subtype of high-grade spindle cell sarcoma, Cold Spring Harbor Molecular Case Studies, Dec. 2018, https://doi.org/10.1101/mcs.a003194 * Dupain C. et al., Discovery of New Fusion Transcripts in a Cohort of Pediatric Solid Cancers at Relapse and Relevance for Personalized Medicine, Molecular Therapy, Jan. 2019, https://doi.org/10.1016/j.ymthe.2018.10.022 * Gu Z. et al., PAX5-driven subtypes of B-progenitor acute lymphoblastic leukemia, Nature Genetics, Jan. 2019, https://dx.doi.org/10.1038/s41588-018-0315-5 * Loarer F.L. et al., Clinicopathologic Features of CIC-NUTM1 Sarcomas, a New Molecular Variant of the Family of CIC-Fused Sarcomas, The American Journal of Surgical Pathology, Feb. 2019, https://dx.doi.org/10.1097/PAS.0000000000001187 * Piarulli G. et al., Gene fusion involving the insulin‐like growth factor 1 receptor in an ALK‐negative inflammatory myofibroblastic tumour, Histopathology, Feb. 2019, https://doi.org/10.1111/his.13839 * Chen S. et al., Widespread and Functional RNA Circularization in Localized Prostate Cancer, Cell, Feb. 2019, https://doi.org/10.1016/j.cell.2019.01.025 * McNeer N.A. et al., Genetic mechanisms of primary chemotherapy resistance in pediatric acute myeloid leukemia, Leukemia, Feb. 2019, https://doi.org/10.1038/s41375-019-0402-3 * Fouchardiere A. et al., B-Catenin nuclear expression discriminates deep penetrating nevi from other cutaneous melanocytic tumors, Virchows Archiv, Feb. 2019, https://doi.org/10.1007/s00428-019-02533-9 * Huang D.W. et al., RNA Sequencing in B-Cell Lymphomas, Lymphoma, Feb. 2019, https://doi.org/10.1007/978-1-4939-9151-8_13 * Heyer E.E. et al., Diagnosis of fusion genes using targeted RNA sequencing, Nature Communications, March 2019, https://doi.org/10.1038/s41467-019-09374-9 * Ravi N. et al., Identification of Targetable Lesions in Anaplastic Thyroid Cancer by Genome Profiling, March 2019, https://doi.org/10.3390/cancers11030402 * Zhu C. et al., The fusion landscape of hepatocellular carcinoma, Molecular Oncology, April 2019, https://doi.org/10.1002/1878-0261.12479 * Schroeder M.P. et al., Integrated analysis of relapsed B-cell precursor Acute Lymphoblastic Leukemia identifies subtype-specific cytokine and metabolic signatures, Scientific Reports, March 2019, https://doi.org/10.1038/s41598-019-40786-1 * Bastian L. et al., PAX5 biallelic genomic alterations define a novel subgroup of B-cell precursor acute lymphoblastic leukemia, Leukemia, March 2019, https://doi.org/10.1038/s41375-019-0430-z * Korshunov A. et al., Desmoplastic/nodular medulloblastomas (DNMB) and medulloblastomas with extensive nodularity (MBEN) disclose similar epigenetic signatures but different transcriptional profiles, Acta Neuropathologica, March 2019, https://doi.org/10.1007/s00401-019-01981-6 * Rampersaud E. et al., Germline deletion of ETV6 in familial acute lymphoblastic leukemia, Blood Advances, April 2019, https://doi.org/10.1182/bloodadvances.2018030635 * Zhang X.C. et al., Comprehensive genomic and immunological characterization of Chinese non-small cell lung cancer patients, Nature Communications, April 2019, http://dx.doi.org/10.1038/s41467-019-09762-1 * Yang W. et al., Immunogenic neoantigens derived from gene fusions stimulate T cell responses, Nature Medicine, Vol. 25, April 201, https://doi.org/10.1038/s41591-019-0434-2 * Frank M.O. et al., Sequencing and curation strategies for identifying candidate glioblastoma treatments, BMC Medical Genomics, April 2019, https://doi.org/10.1186/s12920-019-0500-0 * Yamazaki F. et al., Novel NTRK3 Fusions in Fibrosarcomas of Adults, The American Journal of Surgical Pathology, April 2019, https://doi.org/10.1097/PAS.0000000000001194 * Zhu D. et al., The landscape of chimeric RNAs in bladder urothelial carcinoma, The International Journal of Biochemistry and Cell Biology, May 2019, https://doi.org/10.1016/j.biocel.2019.02.007 * Troll C.J. et al., Structural Variation Detection by Proximity Ligation from Formalin-Fixed, Paraffin-Embedded Tumor Tissue, The Journal of Molecular Diagnostics, May 2019, https://doi.org/10.1016/j.jmoldx.2018.11.003 * Saba K.H. et al., Genetic profiling of a chondroblastoma‐like osteosarcoma/malignant phosphaturic mesenchymal tumor of bone reveals a homozygous deletion of CDKN2A, intragenic deletion of DMD and a targetable FN1‐FGFR1 gene fusion, Genes Chromosomes and Cancer, May 2019, https://doi.org/10.1002/gcc.22764 * Panagopoulos I. et al., Novel GTF2I–PDGFRB and IKZF1–TYW1 fusions in pediatric leukemia with normal karyotype, Experimental Hematology & Oncology, May 2019, https://doi.org/10.1186/s40164-019-0136-y * Trahair T. et al., Crizotinib and Surgery for Long-Term Disease Control in Children and Adolescents With ALK-Positive Inflammatory Myofibroblastic Tumors, JCO Precision Oncology, May 2019, https://doi.org/10.1200/PO.18.00297 * Blackburn J. et al., TMPRSS2‐ERG fusions linked to prostate cancer racial health disparities: A focus on Africa, Prostate, May 2019, https://doi.org/10.1002/pros.23823 * Yang B. et al., Xp11 translocation renal cell carcinoma and clear cell renal cell carcinoma with TFE3 strong positive immunostaining: morphology, immunohistochemistry, and FISH analysis, Modern Pathology, June 2019, https://doi.org/10.1038/s41379-019-0283-z * Hynst J. et al., Bioinformatic pipelines for whole transcriptome sequencing data exploitation in leukemia patients with complex structural variants, PeerJ 7:e7071, June 2019, https://doi.org/10.7717/peerj.7071 * Makise N. et al., Low-grade endometrial stromal sarcoma with a novel MEAF6-SUZ12 fusion**, Virchows Arch., June 2019, https://doi.org/10.1007/s00428-019-02588-8

4 - INSTALLATION AND USAGE EXAMPLES

4.1 - Automatic installation

4.1.1 - Installation using boostrap.py

This is an example of automatic installation of FusionCatcher (and it is installed here “~/fusioncatcher” if these are run in your home directory) and the required databases and indexes (which are downloaded instead of being built locally):

wget http://sf.net/projects/fusioncatcher/files/bootstrap.py -O bootstrap.py && python bootstrap.py -t --download

where: * wget http://sf.net/projects/fusioncatcher/files/bootstrap.py downloads from internet the bootstrap.py which is the installation script (it is recommended to use the boostrap.py from ttp://sf.net/projects/fusioncatcher/files/bootstrap.py because it is more up to date) * python bootstrap.py runs using python the installation script bootstrap.py (here one may replace python with its own custom installation of python, like for example /some/other/custom/python) * -t installs the software tools (and their exact version) needed by FusionCatcher * --download forces the installation script bootstrap.py to download and install automatically also the databases needed by FusionCatcher (if this is not used the databases needed by FusionCatcher will not be installed and the user will have to build/install them manually later)

In case that there are several Python versions installed already then it is possible to point which one to use for installation and running FusionCatcher, as following (no required databases and indexes are installed automatically in this example):

wget http://sf.net/projects/fusioncatcher/files/bootstrap.py -O bootstrap.py

/some/other/python bootstrap.py

In case that one wants to install FusionCatcher here /some/directory/fusioncatcher/, then this shall be run (no required databases and indexes are installed automatically in this example):

wget http://sf.net/projects/fusioncatcher/files/bootstrap.py -O bootstrap.py

/your/favourite/python bootstrap.py --prefix=/some/directory/

In case that one wants to install FusionCatcher and download the databases directly and build locally the indexes, then this shall be run:

wget http://sf.net/projects/fusioncatcher/files/bootstrap.py -O bootstrap.py && python bootstrap.py --build

This is an example of automatic installation of FusionCatcher and the required databases and indexes (which are downloaded instead of being built locally) while all the questions asked by the installation script are answered automatically with YES (WARNING: this might overwrite files/directories):

wget http://sf.net/projects/fusioncatcher/files/bootstrap.py -O bootstrap.py && python bootstrap.py --download -y

In case that one has the admin/root rights then it is possible to install FusionCatcher as following (no required databases and indexes are installed automatically in this example):

wget http://sf.net/projects/fusioncatcher/files/bootstrap.py -O bootstrap.py
sudo python bootstrap.py

In case that one plans not to use at all BLAT with FusionCatcher then add the command line -k to the boostrap.py, as following:

python bootstrap.py -k

In case that you do not know which one to use from these examples, please use the first one! Also, for more info about what options offered by bootstrap.py, please run

bootstrap.py --help

Please, do not forget to build/download the organism data after this is done running (please notice the last lines displayed by bootstrap.py after it finished running and execute the commands suggested there, e.g. use download.sh)!

4.1.2 - Installation using conda

FusionCatcher can be installed also using conda, as follows:

conda config --add channels defaults
conda config --add channels bioconda
conda config --add channels conda-forge
conda create -n fusioncatcher fusioncatcher

After the environment is created, the next steps are:

source activate fusioncatcher
download-human-db.sh

and conda config will permanently add the three channels in the user conda config file. Additionally, creating new and clean environment with conda create is recommended over using conda install.

4.1.3 - Installation from GitHub

git clone https://github.com/ndaniel/fusioncatcher
cd fusioncatcher/tools/
./install_tools.sh
cd ../data
./download-human-db.sh

NOTE: Here it is assumed that Python 2.7.x and BioPython are already installed.

4.2 - Manual installation

This is an example (or one of the many ways) for installing FusionCatcher on a Ubuntu Linux 12.04/14.04 64-bit system and the FusionCatcher and its dependencies are installed in /apps.

• check that Python 2.6.X or 2.7.X is installed and working properly! If not then install it together with its development environment and other (probably) needed dependencies (required):

sudo apt-get install build-essential
sudo apt-get install libncurses5-dev
sudo apt-get install gawk
sudo apt-get install gcc
sudo apt-get install g++
sudo apt-get install make
sudo apt-get install automake
sudo apt-get install gzip
sudo apt-get install bzip2
sudo apt-get install cmake
sudo apt-get install zlib1g-dev
sudo apt-get install zlib1g
sudo apt-get install wget
sudo apt-get install curl
sudo apt-get install pigz
sudo apt-get install zip
sudo apt-get install tar
sudo apt-get install unzip
sudo apt-get install libc6-dev
sudo apt-get install default-jdk
sudo apt-get install libtbb-dev
sudo apt-get install libtbb2
sudo apt-get install parallel
sudo apt-get install python
sudo apt-get install python-dev
sudo apt-get install python-numpy
sudo apt-get install python-biopython
sudo apt-get install python-xlrd
sudo apt-get install python-openpyxl

and for RedHat/CentOS this would be required sudo yum groupinstall "Development Tools" sudo yum install ncurses-devel sudo yum install awk sudo yum install gcc sudo yum install make sudo yum install cmake sudo yum install glibc-devel sudo yum install zlib-devel sudo yum install gzip sudo yum install pigz sudo yum install wget sudo yum install curl sudo yum install tbb-devel sudo yum install python-devel sudo yum install python-biopython sudo yum install python-numpy sudo yum install python-xlrd sudo yum install python-openpyxl sudo yum install java-1.8.0-openjdk* (or other Java?)

and for OpenSUSE this would be required sudo zypper in --type pattern Basis-Devel sudo zypper in gcc sudo zypper in ncurses-devel sudo zypper in python-devel sudo zypper in zlib-devel

• installing BioPython (required):

sudo apt-get install python-numpy
sudo apt-get install python-biopython

• installing Python module xlrd (optional):

sudo apt-get install python-xlrd

• installing Python module openpyxl (optional):

sudo apt-get install python-openpyxl

• create the needed directories:

mkdir -p /apps/fusioncatcher/tools
mkdir -p /apps/fusioncatcher/data

• installing Bowtie 64-bit version 1.2.0 (required)

cd /apps/fusioncatcher/tools
wget https://github.com/BenLangmead/bowtie/releases/download/v1.2.3/bowtie-1.2.3-linux-x86_64.zip
unzip bowtie-1.2.3-linux-x86_64.zip
ln -s bowtie-1.2.3-linux-x86_64 bowtie

• installing Bowtie2 64-bit version 2.3.5.1 (required)

cd /apps/fusioncatcher/tools
wget https://github.com/BenLangmead/bowtie2/releases/download/v2.3.5.1/bowtie2-2.3.5.1-linux-x86_64.zip
unzip bowtie2-2.3.5.1-linux-x86_64.zip
ln -s bowtie2-2.3.5.1-linux-x86_64 bowtie2

• installing BLAT version 0.35 (optional; if BLAT is not installed please use option ‘–skip-blat’ or remove the blat string from the aligners =... line of file fusioncatcher/etc/configuration.cfg in order to let know FusionCatcher that it should not use it)

cd /apps/fusioncatcher/tools
mkdir blat_v0.35
cd blat_v0.35
wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/blat/blat
chmod +x blat
wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/faToTwoBit
chmod +x faToTwoBit
wget http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64/liftOver
chmod +x liftOver
cd ..
ln -s blat_v0.35 blat

• installing fastq-dump version 2.6.2 from NCBI SRA Toolkit version 2.6.2 (optional but it is required in case that one wants to test the installation using the example given below)

cd /apps/fusioncatcher/tools
wget http://ftp-private.ncbi.nlm.nih.gov/sra/sdk/2.9.6/sratoolkit.2.9.6-centos_linux64.tar.gz
tar zxvf sratoolkit.2.9.6-centos_linux64.tar.gz
ln -s sratoolkit.2.9.6-centos_linux64 sratoolkit

• installing SeqTK version 1.0-r82b (please note that this is a different development branch than the official development) (required)

cd /apps/fusioncatcher/tools
wget http://github.com/ndaniel/seqtk/archive/1.0-r82b.tar.gz -O 1.0-r82b.tar.gz
tar zxvf 1.0-r82b.tar.gz
cd seqtk-1.0-r82b
make
cd ..
ln -s seqtk-1.0-r82b seqtk

• installing STAR version 2.7.2b (required)

cd /apps/fusioncatcher/tools
wget http://github.com/alexdobin/STAR/archive/2.7.2b.tar.gz -O 2.7.2b.tar.gz
tar zxvf 2.7.2b.tar.gz
cd 2.7.2b
cd source
rm -f STAR
cp ../bin/Linux_x86_64_static/STAR .

Try to run this command (if it fails please ignore the error messages and continue further; continue further either way) make

and continue with cd .. ln -s 2.7.2b star

• installing Velvet version 1.2.10 (optional)

cd /apps/fusioncatcher/tools
wget http://www.ebi.ac.uk/~zerbino/velvet/velvet_1.2.10.tgz
tar zxvf velvet_1.2.10.tgz
cd velvet_1.2.10
make
cd ..
ln -s velvet_1.2.10 velvet

Note: Velvet depends on zlib-dev which may be installed like this

sudo apt-get install zlib-dev

• installing coreutils version 8.25 for a newer version of sort command which allows use of several CPUs in parallel, that is the use of --parallel command line option (optional)

cd /apps/fusioncatcher/tools
wget http://ftp.gnu.org/gnu/coreutils/coreutils-8.25.tar.xz
tar --xz -xvf coreutils-8.25.tar.xz
cd coreutils-8.25
./configure
make
cd ..
ln -s coreutils-8.25 coreutils

• installing pigz version 2.4 (optional)

cd /apps/fusioncatcher/tools
wget http://zlib.net/pigz/pigz-2.4.tar.gz
tar zxvf pigz-2.3.1.tar.gz
cd pigz-2.4
make
cd ..
ln -s pigz-2.4 pigz

• installing Picard tools version 2.2.2 (optional)

cd /apps/fusioncatcher/tools
mkdir picard
cd picard
wget http://github.com/broadinstitute/picard/releases/download/2.21.2/picard.jar
cd ..

• installing FusionCatcher version 1.20 (required)

cd /apps/fusioncatcher
wget http://sourceforge.net/projects/fusioncatcher/files/fusioncatcher_v1.20.zip
unzip fusioncatcher_v1.20.zip
cd fusioncatcher_v1.20

rm -rf ../bin
rm -rf ../etc
rm -rf ../doc
rm -rf ../VERSION
rm -rf ../NEWS
rm -rf ../LICENSE
rm -rf ../README.md
rm -rf ../DEPENDENCIES

ln -s $(pwd)/bin ../bin
ln -s $(pwd)/etc ../etc
ln -s $(pwd)/doc ../doc
ln -s $(pwd)/test ../test
ln -s $(pwd)/VERSION ../VERSION
ln -s $(pwd)/NEWS ../NEWS
ln -s $(pwd)/LICENSE ../LICENSE
ln -s $(pwd)/README.md ../README.md
ln -s $(pwd)/DEPENDENCIES ../DEPENDENCIES

• specify the paths to the above tools such that FusionCatcher can find them. There are two choices.
• Choice A: Edit the FusionCatcher configuration file configuration.cfg (type: nano /apps/fusioncatcher/etc/configuration.cfg at command line), which has priority over PATH, and make sure that the FusionCatcher’s configuration file ‘configuration.cfg’ looks like this:

[paths]
python = /usr/bin/
data = /apps/fusioncatcher/data/current/
scripts = /apps/fusioncatcher/bin/
bowtie = /apps/fusioncatcher/tools/bowtie/
blat = /apps/fusioncatcher/tools/blat/
bowtie2 = /apps/fusioncatcher/tools/bowtie2/
star = /apps/fusioncatcher/tools/star/source/
seqtk = /apps/fusioncatcher/tools/seqtk/
velvet = /apps/fusioncatcher/tools/velvet/
fatotwobit = /apps/fusioncatcher/tools/blat/
liftover = /apps/fusioncatcher/tools/blat/
sra = /apps/fusioncatcher/tools/sratoolkit/bin/
numpy = /apps/fusioncatcher/tools/numpy/
biopython = /apps/fusioncatcher/tools/biopython/
xlrd = /apps/fusioncatcher/tools/xlrd/
openpyxl = /apps/fusioncatcher/tools/openpyxl
lzop = /apps/fusioncatcher/tools/lzop/src/
coreutils = /apps/fusioncatcher/tools/coreutils/src/
pigz = /apps/fusioncatcher/tools/pigz/
samtools = /apps/fusioncatcher/tools/samtools/
picard = /apps/fusioncatcher/tools/picard/
parallel = /appsfusioncatcher/tools/paralell/src/
bbmap = /apps/fusioncatcher/tools/bbmap/
pxz = /apps/fusioncatcher/tools/pxz/
java = /usr/bin/
[parameters]
threads = 0
aligners = blat,star
[versions]
fusioncatcher = 1.20

• Choice B: Add the paths for the needed tools to the PATH variable by editing, for example, the .bashrc file (type: nano ~/.bashrc at command line) and add the following lines at the end:

export PATH=/apps/fusioncatcher/bin:$PATH
export PATH=/apps/fusioncatcher/tools/bowtie:$PATH
export PATH=/apps/fusioncatcher/tools/bowtie2:$PATH
export PATH=/apps/fusioncatcher/tools/blat:$PATH
export PATH=/apps/fusioncatcher/tools/star/source/:$PATH
export PATH=/apps/fusioncatcher/tools/liftover:$PATH
export PATH=/apps/fusioncatcher/tools/seqtk:$PATH
export PATH=/apps/fusioncatcher/tools/sratoolkit/bin:$PATH
export PATH=/apps/fusioncatcher/tools/velvet/:$PATH
export PATH=/apps/fusioncatcher/tools/fatotwobit/:$PATH
export PATH=/apps/fusioncatcher/tools/lzop/src/:$PATH
export PATH=/apps/fusioncatcher/tools/coreutils/src/:$PATH
export PATH=/apps/fusioncatcher/tools/pigz/:$PATH
export PATH=/apps/fusioncatcher/tools/samtools/:$PATH
export PATH=/apps/fusioncatcher/tools/bbmap/:$PATH
export PATH=/apps/fusioncatcher/tools/picard/:$PATH

Note 1: If a different version of Python is used/needed by FusionCatcher than the standard /usr/bin/env python then also please make sure that that specific version of Python is added to the PATH variable by editing, for example, the .bashrc file (type: nano ~/.bashrc at command line) or add the following lines at the end:

export PATH=/some/other/version/of/python:$PATH

Note 2: fusioncatcher/etc/configuration.cfg has priority over $PATH.

Note 3: In some cases it might not be enough to change the Python’s path in .bashrc file, like for example the case when FusionCatcher is run on a server which defaults to another Python than one used to install FusionCatcher. In this case it is required that one changes all the shebangs of the all Python scripts which belong to FusionCatcher. In case that one uses the Python which has the following Python executable path /some/other/python than this can be done like this (it changes in place /usr/bin/env python into /some/other/python in all /apps/fusioncatcher/bin/*.py):

sed -i 's/\/usr\/bin\/env\ python/\/some\/other\/python/g' /apps/fusioncatcher/bin/*.py

• download/build the human build data from Ensembl database and other databases and build the necessary indexes and files (the latest release of Ensembl data is release 95 as March 2019 when this section was updated last time). There are two alternative ways to get the human build data. The recommended way is to use fusioncatcher-build.
• Using direct download

mkdir -p /apps/fusioncatcher/data
cd /apps/fusioncatcher/data
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.aa
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.ab
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.ac
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.ad
cat human_v98.tar.gz.* | tar xz
ln -s human_v98 current

• Using fusioncatcher-build – It will takes several hours (e.g. 5-10 hours) and it depends highly on the bandwidth of your internet connection. One may find out what Ensembl database version is available at [www.ensembl.org] and what version has been downloaded by looking to the last three lines printed on the screen by fusioncatcher-build.

mkdir -p /apps/fusioncatcher/data/human_v98
cd /apps/fusioncatcher/data/human_v98
/apps/fusioncatcher/bin/fusioncatcher-build -g homo_sapiens -o .
cd ..
ln -s human_v98 current

4.3 - Semi-automatic installation

This is an example of semi-automatic installation of FusionCatcher (and it is installed here: /some/server/apps/fusioncatcher). This may be used when FusionCatcher should be installed on a computer without internet connection. Shortly, in this case all the software dependencies and indexes of databases need to be downloaded separately on another computer which has internet connection and from there they should be copied/moved to the computer without internet connection. Here are the steps for achieving these:

• on computer A (which has internet connection):
• create locally a folder named, for example, fuscat: mkdir fuscat cd fuscat
• download bootstrap.py wget http://sf.net/projects/fusioncatcher/files/bootstrap.py
• find out the dependencies needed to be downloaded and download them manually into folder fuscat (their URLs will be shown and the user needs to download them manually using wget or its favourite browser) python bootstrap.py --list-dependencies ...
• copy/move (manually) the folder fuscat and all its content to computer B (which does not have internet connection) where one intends to install FusionCatcher
• on the computer B (which does not have internet connection), where one intends to install FusionCatcher:
• go to the folder fuscat and make sure that the downloaded files do not have their permissions set as executables (this might confuse bootstrap.py) cd fuscat chmod -x *
• start the installing process of FusionCatcher using bootstrap.py (if one wishes to use another version of Python, like for example having the path /some/other/python then below please replace python with /some/other/python) python bootstrap.py --local .
• for installing the pre-built index files for human, please run (or take a look for instructions to) download.sh (it should be in bin directory where FusionCatcher has been installed)
• for installing the pre-built index files for other organisms than human please, use fusioncatcher-build according tot the manual

For more information regarding the installation settings and possibilities, run:

python bootstrap.py --help

4.4 - Testing installation

This test works only when human organism.

4.4.1 - Automatic

Here are the steps for testing the installation of FusionCatcher using human genome.

cd ~
/apps/fusioncatcher/test/test.sh

Afterwards a message will be shown at console if the installation test went fine or not.

4.4.1 - Manual

Here are the steps for testing the installation of FusionCatcher using human genome.

mkdir ~/test
cd ~/test

wget http://sourceforge.net/projects/fusioncatcher/files/test/reads_1.fq.gz
wget http://sourceforge.net/projects/fusioncatcher/files/test/reads_2.fq.gz

cd ..

/apps/fusioncatcher/bin/fusioncatcher \
-d /apps/fusioncatcher/data/current/ \
--input ~/test/ \
--output ~/test-results/

This should take around 5 minutes to run and the result file ~/test-results/final-list_candidates-fusion-genes.txt should look like this.

This dataset contains a very small set of short reads covering 12 already known fusion genes from human tumor cell lines which have been RNA sequenced (for more see here).

4.5 - Breast cancer cell line

This is an example of finding fusion genes in the BT474 cell line using the public available RNA-seq data (from SRA archive): * download the publicly available RNA-seq data for BT-474 tumor breast cell line published in article H. Edgren, A. Murumagi, S. Kangaspeska, D. Nicorici, V. Hongisto, K. Kleivi, I.H. Rye, S. Nyberg, M. Wolf, A.L. Borresen-Dale, O.P. Kallioniemi, Identification of fusion genes in breast cancer by paired-end RNA-sequencing, Genome Biology, Vol. 12, Jan. 2011 http://genomebiology.com/2011/12/1/R6/ : mkdir -p ~/bt474 cd ~/bt474 wget http://ftp-private.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/SRR064/SRR064438/SRR064438.sra wget http://ftp-private.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/SRR064/SRR064439/SRR064439.sra

• run FusionCatcher (it takes around ~2.5 hours):

/apps/fusioncatcher/bin/fusioncatcher \
-d /apps/fusioncatcher/data/current/ \
-i ~/bt474/ \
-o ~/bt474_fusions/

• if the run was successful then there should be the (non-empty) files (for more information see here):

~/bt474_fusions/final-list_candidate_fusion_genes.txt
~/bt474_fusions/summary_candidate_fusions.txt
~/bt474_fusions/viruses_bacteria_phages.txt
~/bt474_fusions/supporting-reads_gene-fusions_BOWTIE.zip
~/bt474_fusions/supporting-reads_gene-fusions_BLAT.zip
~/bt474_fusions/supporting-reads_gene-fusions_STAR.zip
~/bt474_fusions/info.txt
~/bt474_fusions/fusioncatcher.log

and the file ~/bt474_fusions/final-list_candidate_fusion_genes.txt

should contain almost all fusion genes which have been published here: * S. Kangaspeska, S. Hultsch, H. Edgren, D. Nicorici, A. Murumägi, O.P. Kallioniemi, Reanalysis of RNA-sequencing data reveals several additional fusion genes with multiple isoforms, PLOS One, Oct. 2012. http://dx.plos.org/10.1371/journal.pone.0048745 * H. Edgren, A. Murumagi, S. Kangaspeska, D. Nicorici, V. Hongisto, K. Kleivi, I.H. Rye, S. Nyberg, M. Wolf, A.L. Borresen-Dale, O.P. Kallioniemi, Identification of fusion genes in breast cancer by paired-end RNA-sequencing, Genome Biology, Vol. 12, Jan. 2011. http://genomebiology.com/2011/12/1/R6

4.6 - Batch mode

This is an example of finding fusion genes in the Illumina Body Map 2.0 RNA-seq data which consists of 16 RNA samples from 16 different organs from healthy persons. For doing this the batch mode is used (where the input is this file), as shown here:

mkdir -p ~/bodymap
cd ~/bodymap
wget http://sourceforge.net/projects/fusioncatcher/files/examples/illumina-bodymap2.txt
fusioncatcher-batch.py -i illumina-bodymap2.txt -o results

The input file for fusioncatcher-batch.py is a text tab-separated file with two columns and 16 lines (one line for each organ from Illumina Body Map 2.0). The first column contains the URLs for the input FASTQ files and the second column (which is optional) contains the name of the organ (which will be used to create a output directory later where the results will be). Therefore, FusionCatcher will be run automatically 16 times by the fusioncatcher-batch.py.

The fusion genes found in Illumina Body Map 2.0 could be used later, for example, as a list of known false positives when looking for fusion genes in diseased/tumor samples.

4.7 - Matched normal sample

In case that there is available RNA-seq data from a tumor sample and its match normal sample then the somatic mode of FusionCatcher may be used. By default FusionCatcher is using a background list of fusion genes which have been found previously in normal healthy samples (e.g. Illumina BodyMap2 , etc.).

For example, lets assume that (i) the BT-474 is the rumor sample from here, and (ii) the matched normal samples if the healthy breast sample from here. In this case, in order to find the somatic fusion genes in the BT-474 (that are the fusion genes which are found in BT-474 and are not found in the healthy sample) FusionCatcher should be run as follows:

mkdir -p ~/bt474
cd ~/bt474
wget http://ftp-private.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/SRR064/SRR064438/SRR064438.sra
wget http://ftp-private.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/SRR064/SRR064439/SRR064439.sra


mkdir -p ~/healthy
cd ~/healthy
wget http://ftp-trace.ncbi.nlm.nih.gov/sra/sra-instant/reads/ByRun/sra/SRR/SRR064/SRR064437/SRR064437.sra

cd ~

fusioncatcher.py
--input ~/bt474/
--normal ~/healthy
--output ~/results/

The somatic fusion genes for BT-474 will be found in ~/results/bt474/final-list_candidate_fusion_genes.txt file. The fusion genes which marked as matched-normal (see column 3) in BT-474 (that is ~/results/bt474/final-list_candidate_fusion_genes.txt file) have been found also in the healthy sample also and most likely they are not somatic.

In case that there are several tumor samples and their matched healthy samples then batch mode of FusionCatcher may be used, as follows:

fusioncatcher-batch.py
--input /some/path/tumor-file.txt
--normal /some/path/healthy-file.txt
--output /some/path/results/

where: * /some/path/tumor-file.txt is a text file containing on each line a path to FASTQ files belonging to the tumor cells (an example is here), * /some/path/healthy-file.txt is a text file containing on each line a path to FASTQ files belonging to the healthy cells (an example is here), * /some/path/results is the output directory where the results are placed.

4.8 - Edgren RNA-seq dataset

This is an example of finding fusion genes in the Edgren RNA-seq data (from SRA archive): * H. Edgren, A. Murumagi, S. Kangaspeska, D. Nicorici, V. Hongisto, K. Kleivi, I.H. Rye, S. Nyberg, M. Wolf, A.L. Borresen-Dale, O.P. Kallioniemi, Identification of fusion genes in breast cancer by paired-end RNA-sequencing, Genome Biology, Vol. 12, Jan. 2011. http://genomebiology.com/2011/12/1/R6 * S. Kangaspeska, S. Hultsch, H. Edgren, D. Nicorici, A. Murumägi, O.P. Kallioniemi, Reanalysis of RNA-sequencing data reveals several additional fusion genes with multiple isoforms, PLOS One, Oct. 2012. http://dx.plos.org/10.1371/journal.pone.0048745

fusioncatcher-batch.py -i http://sourceforge.net/projects/fusioncatcher/files/examples/edgren.txt -o results/

NOTE: DO NOT POOL the samples from all these cell lines. DO NOT give at once all these SRA/FASTQ files as input to FusionCatcher! Run FusionCatcher separately for each cell line! It is ok the pool together the samples from the same cell line together (but still do not concatenate yourself the FASTQ files and let FusionCatcher do it for you)!

5 - QUICK INSTALLATION

5.1 - Getting executables

For a fully automatic installation (including the required indexes of databases) run:

wget http://sf.net/projects/fusioncatcher/files/bootstrap.py && python bootstrap.py --download

In case of a manual installation, first please check that (i) the required dependencies are installed, and (ii) download the source files of FusionCatcher, like for example:

wget http://sourceforge.net/projects/fusioncatcher/files/fusioncatcher_v1.20.zip 
unzip fusioncatcher_v1.20.zip

For an example of: * fully automatic installation see here, * manual installation see here, and * semi-automatic installation see here.

5.2 - Organism’s build data

First, it is needed to download data or build the necessary files/indexes for running the FusionCatcher. This process should be done once for every single organism or every time when the Ensembl database is updated.

5.2.1 - Direct download of human build data

Here, in this example, the necessary data is downloaded and necessary files/indexes for the human genome are downloaded in the directory /some/human/data/human_v98/ which will be used later.

mkdir -p /some/human/data/
cd /some/human/data/
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.aa
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.ab
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.ac
wget http://sourceforge.net/projects/fusioncatcher/files/data/human_v98.tar.gz.ad
cat human_v98.tar.gz.* | tar xz
ln -s human_v98 current

If this works then it is not necessary to start building yourself the build data as shown below (which is only needed in case that the direct download for some reason does not work or one wishes to use the build data of another organism which is not available for download).

5.2.2 - Building yourself the organism’s build data

Here, in this example, the necessary data is downloaded and necessary files/indexes are built for the human genome in the directory /some/human/data/directory/ which will be used later.

fusioncatcher-build -g homo_sapiens -o /some/human/data/directory/

This takes around 5-10 hours (downloading, building indexes/databases, etc.).

In case that one wants to use a Ensembl server which is situated geographically closer, then one has: * Ensembl server in Europe (used by default):

fusioncatcher-build -g homo_sapiens -o /some/human/data/directory/

• Ensembl server in East US:

fusioncatcher-build -g homo_sapiens -o /some/human/data/directory/ --web=useast.ensembl.org

• Ensembl server in West US:

fusioncatcher-build -g homo_sapiens -o /some/human/data/directory/ --web=uswest.ensembl.org

• Ensembl server in Asia:

fusioncatcher-build -g homo_sapiens -o /some/human/data/directory/ --web=asia.ensembl.org

In case, that it is not possible to use fusioncatcher-build for vary reasons (e.g. access to Ensembl website is very slow) then one may directly download the latest human build data (generated by fusioncatcher-build using Ensembl database release 95) necessary for running FusionCatcher from here (all files are needed and the total size is ~25 GB).

For rat genome, one has

fusioncatcher-build -g rattus_norvegicus -o /some/rat/data/directory/

For mouse genome, one has

fusioncatcher-build -g mus_musculus -o /some/mouse/data/directory/

NOTE: FusionCatcher version 1.20 needs a newer build data than the previous version (that is 1.10) of ‘fusioncatcher-build’.

6 - USAGE

Searching for fusion genes in a human organism, one has:

fusioncatcher \
-d /some/human/data/directory/ \
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/

where: * /some/human/data/directory/ - contains the data and files generated by fusioncatcher-build (see Get data section) * /some/input/directory/containing/fastq/files/ - contains the input FASTQ (or SRA if NCBI SRA toolkit is installed) files (and not any other type of files which are not do not contain sequecing data, e.g. readme.txt) * /some/output/directory/ - contains output files (for more information see here): * final-list_candidate_fusion_genes.txt * summary_candidate_fusions.txt * supporting-reads_gene-fusions_BOWTIE.zip * supporting-reads_gene-fusions_BLAT.zip * supporting-reads_gene-fusions_STAR.zip * viruses_bacteria_phages.txt * info.txt * fusioncatcher.log

Searching for fusion genes in a rat organism, one has:

fusioncatcher \
-d /some/rat/data/directory/ \
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/

Searching for fusion genes in a mouse organism, one has:

fusioncatcher \
-d /some/mouse/data/directory/ \
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/

6.1 - Input data

The input data shall be: * a directory containing the input FASTQ/SRA files (this is highly recommended), or * several files separated by comma, e.g. file01.fq,file02.fq (there should be no blank before and after the comma!).

All types of raw FASTQ files produced by the Illumina Solexa and Illumina HiSeq platforms containing paired-end reads may be given as input. The FASTQ files shall: * come from an RNA sequencing experiment (i.e. the transcriptome/RNA is sequenced), and * contain paired-end reads, and * paired FASTQ files should be synchronized (i.e. reads which for a pair should be on the same line number if both FASTQ files). * paired-end reads which follow the suggested Illumina’s sample preparation protocol, that is the two read-mates are: (i) from opposite strands, and (ii) opposite directions to one another (in other words, in order to ‘bring’ a read and its mate-read on the same strand then one needs to perform reverse-complement operation on only one of them) * paired-end reads shall come from a stranded (i.e. strand-specific) or unstranded sample preparation protocol (both are supported by FusionCatcher)!

It is highly recommended that: * the input FASTQ files contain the raw reads generated by the Illumina sequencers without any additional trimming (i.e. all reads from all files shall have the same length), and * every single input FASTQ file contains reads from only and only one sample/replicate (i.e. do not concatenate in one big FASTQ file several other FASTQ files; just give the input all FASTQ files and FusionCatcher will do the concatenation).

FusionCatcher will automatically pre-process the input reads, as follows: * trimming 3’ end of the reads based on quality scores (default Q5), * removing automatically the adapter from the reads (it predicts the adapter sequence based on the reads which form a pair and also overlap and the non-overlapping parts are the predicted adapters), * trimming the poly A/C/G/T tails, * removing the reads which contain short tandem repeats (see: M. Gymrek, et al. lobSTR: A short tandem repeat profiler for personal genomes, Genome Res. 2012 Jun;22(6):1154-62, here http://genome.cshlp.org/content/22/6/1154.abstract , * removing the reads which are marked as bad by Illumina sequencer, * removing the reads which are too short after the trimming, * removing the reads which map on ribosomal RNA, * removing the reads which map on genomes of bacteria/phages/viruses (from: ftp://ftp.ncbi.nlm.nih.gov/genomes/Viruses/ ).

Note: If the reads contains barcode sequences then they should be removed before given as inputs to FusionCatcher.

The SRA files are accepted as input as long as the NCBI SRA toolkit (see dependencies section) is installed and available.

The following files are accepted/used as input: * *.fq.zip * *.fq.gz * *.fq.bz2 * *.fastq.zip * *.fastq.gz * *.fastq.bz2 * *.txt.zip * *.txt.gz * *.txt.bz2 * *.fq * *.fastq * *.sra and the zip and gz archives should contain only one file.

FusionCatcher also accepts as input also URLs (it shall start with ftp:// or http://), like for example

fusioncatcher -i ftp://ftp.sra.ebi.ac.uk/vol1/fastq/ERR030/ERR030872/ERR030872_1.fastq.gz,ftp://ftp.sra.ebi.ac.uk/vol1/fastq/ERR030/ERR030872/ERR030872_2.fastq.gz -o thyroid

The FASTQ input files should be the ones generated by the Illumina platform without any kind of additional processing/filtering/trimming. FusionCatcher is doing its own filtering (automatically identifies and trims the adapters, trimming if needed, quality filtering of reads, removal of rRNA reads, etc.).

FusionCatcher is using the file names of the input files in order to figure out which files are paired (which contains ´R1´ reads and which contains the corresponding ´R2´ reads). For this FusionCatcher is ordering alphabetically the file names (found in a input directory) and it considers that the first two form a pair, the next two are forming a pair and so on. Shortly, here the first two files, the next two files and so on should be synchronized. For example, considering the following input files:

L002_R1.fastq.gz
L002_R2.fastq.gz
L003_R1.fastq.gz
L003_R2.fastq.gz

FusionCatcher would automatically figure out correctly that the first two files form a pair and the next two would form another pair.

There are cases when ordering alphabetically the input file names would make FusionCatcher to pair wrongly the input files (i.e. the input FASTQ files are not synchronized). In this kind of cases, renaming the input files such that they fit this rule helps. For example, considering the following input files (where the file containing ´R1´ reads is split into two files and the file containing ´R2´ reads is split into another two files):

L002_R1_01.fastq.gz
L002_R1_02.fastq.gz
L002_R2_01.fastq.gz
L002_R2_02.fastq.gz

FusionCatcher would automatically figure out wrongly that the first two files form a pair and the next two would form another pair. In this case, renaming the files like this

mv L002_R1_01.fastq.gz 01_L002_R1_01.fastq.gz 
mv L002_R1_02.fastq.gz 03_L002_R1_02.fastq.gz 
mv L002_R2_01.fastq.gz 02_L002_R2_01.fastq.gz 
mv L002_R2_02.fastq.gz 04_L002_R2_02.fastq.gz 

would give this alphabetically order list:

01_L002_R1_01.fastq.gz
02_L002_R2_01.fastq.gz
03_L002_R1_02.fastq.gz
04_L002_R2_02.fastq.gz

and with this input files FusionCatcher would work correctly. Another way, around this would be to give the input files separated by comma (in the correct order and no blanks before and after the comma), like this

fusioncatcher -i L002_R1_01.fastq.gz,L002_R2_01.fastq.gz,L002_R1_02.fastq.gz,L002_R2_02.fastq.gz 

For example, this is a valid input:

01_L002_R1_01.fastq.gz
02_L002_R2_01.fastq.gz
03_L002_R1_02.fastq.gz
04_L002_R2_02.fastq.gz
05_L002_R1_03.fastq.gz
06_L002_R2_03.fastq.gz
07_L002_R1_04.fastq.gz
08_L002_R2_04.fastq.gz
09_L002_R1_05.fastq.gz
10_L002_R2_05.fastq.gz
11_L002_R1_06.fastq.gz
12_L002_R2_06.fastq.gz

For example, this is NOT a valid input:

10_L002_R2_05.fastq.gz
11_L002_R1_06.fastq.gz
12_L002_R2_06.fastq.gz
1_L002_R1_01.fastq.gz
2_L002_R2_01.fastq.gz
3_L002_R1_02.fastq.gz
4_L002_R2_02.fastq.gz
5_L002_R1_03.fastq.gz
6_L002_R2_03.fastq.gz
7_L002_R1_04.fastq.gz
8_L002_R2_04.fastq.gz
9_L002_R1_05.fastq.gz

NOTE: * In case that a directory is given as input, one shall make sure that the input directory does not contain files which do not contain reads sequences (e.g. readme.txt, info.txt, etc.)! * Please, let FusionCatcher do the the concatenation of several FASTQ files (i.e. just put all the FASTQ files into one folder and give that folder as input to FusionCatcher) and do NOT do concatenate the FASTQ files yourself (e.g. using cat). This is because most likely different FASTQ files might have: * different adapter sequences (FusionCatcher is expecting that there are only one type of adapter, exactly like it comes directly from the Illumina sequencers), * FusionCatcher does not support the FASTQ files where the reads’ sequences contain barcode sequences (the barcode sequences shoud be removed/trimmed first and only then the trimmed FASTQ file should be given as input to FusionCatcher) * different fragment sizes, and * different read lengths. * DO NOT POOL samples from different cell lines or from different patients! Run FusionCatcher separately with one sample at the time! It is ok the pool together the samples, which come from the (i) same cell line, or (ii) from the same patient (but still do not concatenate yourself the FASTQ files and let FusionCatcher do it for you)!

6.2 - Output data

FusionCatcher produces a list of candidate fusion genes using the given input data. It is recommended that this list of candidate of fusion genes is further validated in the wet-lab using for example PCR/FISH experiments.

The output files are: * final-list_candidate_fusion_genes.txt - final list with the newly found candidates fusion genes (it contains the fusion genes with their junction sequence and points); Starting with version 0.99.3c the coordinates of fusion genes are given here for human genome using only assembly hg38/GRCh38; See Table 1 for columns’ descriptions; * final-list_candidate_fusion_genes.hg19.txt - final list with the newly found candidates fusion genes (it contains the fusion genes with their junction sequence and points); Starting with version 0.99.3d the coordinates of fusion genes are given here for human genome using assembly hg19/GRCh37; See Table 1 for columns’ descriptions; * summary_candidate_fusions.txt - contains an executive summary (meant to be read directly by the medical doctors or biologist) of candidate fusion genes found; * final-list_candidate_fusion_genes.caption.md.txt - explains in detail the labels found in column Fusion_description of files final-list_candidate_fusion_genes.txt and final-list_candidate_fusion_genes.hg19.txt; * supporting-reads_gene-fusions_BOWTIE.zip - sequences of short reads supporting the newly found candidate fusion genes found using only and exclusively the Bowtie aligner; * supporting-reads_gene-fusions_BLAT.zip - sequences of short reads supporting the newly found candidate fusion genes found using Bowtie and Blat aligners; * supporting-reads_gene-fusions_STAR.zip - sequences of short reads supporting the newly found candidate fusion genes found using Bowtie and STAR aligners; * supporting-reads_gene-fusions_BOWTIE2.zip - sequences of short reads supporting the newly found candidate fusion genes found using Bowtie and Bowtie2 aligners; * viruses_bacteria_phages.txt - (non-zero) reads counts for each virus/bacteria/phage from NCBI database ftp://ftp.ncbi.nlm.nih.gov/genomes/Viruses/ * info.txt - information regarding genome version, Ensembl database version, versions of tools used, read counts, etc.; * fusioncatcher.log - log of the entire run (e.g. all commands/programs which have been run, command line arguments used, running time for each command, etc.).

FusionCatcher reports: * multiple times (up to four times) exactly the same candidate fusion gene, which has exactly the same fusion points/junction (i.e. FusionCatcher will output separately the fusions found for each of its four aligners/methods such that it is easy to see what method was used to find a fusion gene) * reciprocal fusion genes if they are found (e.g. geneA-geneB and also geneB-geneA) * every alternative splicing event found for each fusion gene (i.e. alternative fusion isoforms of the same fusion gene)

Table 1 - Columns description for file final-list_candidate-fusion-genes.txt

Table 2 - Labels used to describe the found fusion genes (column Fusion_ description from file final-list_candidate-fusion-genes.txt)

6.3 - Visualization

FusionCatcher outputs also the zipped FASTA files containing the reads which support the found candidate fusions genes. The files are: * supporting-reads_gene-fusions_BOWTIE.zip, * supporting-reads_gene-fusions_BLAT.zip, * supporting-reads_gene-fusions_STAR.zip, * supporting-reads_gene-fusions_BOWTIE2.zip.

The reads which support the: * junction of the candidate fusion have their name ending with _supports_fusion_junction, and * candidate fusion (i.e. one reads map on one gene and the paired-read maps on the other fusion gene) have their name ending with _supports_fusion_pair.

These supporting reads (given as FASTA and FASTQ files) may be used for further visualization purposes. For example, one may use these supporting reads and align them himself/herself using his/her favourite: * aligner (e.g. Bowtie/Bowtie2/TopHat/STAR/GSNAP/etc.), * version/assembly of genome, * mapping format output (e.g. SAM/BAM), and * NGS visualizer (e.g. IGV/UCSC Genome Browser/etc.)

6.3.1 - UCSC Genome Browser

For example, the sequences of supporting reads for a given candidate fusion gene may be visualized using UCSC Genome Browser by aligning them using the UCSC Genome Browser’s BLAT aligner (i.e. copy and paste the reads here: BLAT tool of UCSC Genome Browser –> click the button Submit –> navigate into the UCSC Genome Browser to the genes that form the fusion genes). Also zooming out several times gives better view here.

6.3.2 - PSL format

If one uses the --visualization-psl command line option of the FusionCatcher then the BLAT alignment of the supporting reads will be done automatically by the FusionCatcher and the results are saved in PSL format files with names that are ending with _reads.psl in the: * supporting-reads_gene-fusions_BOWTIE.zip, * supporting-reads_gene-fusions_BLAT.zip, * supporting-reads_gene-fusions_STAR.zip, and * supporting-reads_gene-fusions_BOWTIE2.zip.

The files with names ending in _reads.psl may be used further for visualization of the candidate fusion genes using UCSC Genome Browser, IGV (Integrative Genome Viewer) or any other viewer/browser which supports the PSL format.

Note: If one generated the build files using fusioncatcher-build.py the command line --visualization-psl option should work just fine. If one downloaded the build files then the command line option --visualization-psl will not work an it needs to be enabled by creating manually first the file fusioncatcher/data/current/genome.2bit for FusionCatcher, something like this (here the assumption is that the build files for one’s organism of interest are in fusioncatcher/data/current/):

# re-build the genome index using BLAT where the genome is given FASTA file genome.fa
fusioncatcher/tools/bowtie/bowtie-inspect fusioncatcher/data/current/genome_index/ > fusioncatcher/data/current/genome.fa
fusioncatcher/tools/blat/faToTwoBit fusioncatcher/data/current/genome.fa fusioncatcher/data/current/genome.2bit -noMask

6.3.3 - SAM format

6.3.3.1 - Automatic method

If one uses the --visualization-sam command line option of the FusionCatcher then the BOWTIE2 alignment of the supporting reads will be done automatically by the FusionCatcher and the results are saved as SAM files with names that are ending with _reads.sam in the: * supporting-reads_gene-fusions_BOWTIE.zip, * supporting-reads_gene-fusions_BLAT.zip, * supporting-reads_gene-fusions_STAR.zip, * supporting-reads_gene-fusions_BOWTIE2.zip.

The files with names ending in _reads.sam (please note, that they still needed to be converted to BAM, coordiante sorted and indexed first) may be used further for visualization of the candidate fusion genes using UCSC Genome Browser, IGV (Integrative Genome Viewer) or any other viewer/browser which supports the SAM format.

6.3.3.2 - Manual method

Here is an rough example of manually aligning the supporting reads (that is named as supporting_reads.fq in the below example; the FASTQ files needed here are the files ending in _reads.fq from the ZIP archives supporting-reads_gene-fusions_*.zip produced by FusionCatcher) using different aligners. * Bowtie2 aligner (where your_choice_of_genome_bowtie2_index may be for human, for example this):

# alignment done ignoring the paired-end information (i.e. like single reads):

bowtie2 \
--local \
-k 10 \
-x your_choice_of_genome_bowtie2_index \
-U supporting_reads.fq \
-S fusion_genes.sam

samtools view -bS fusion_genes.sam | samtools sort - fusion_genes.sorted

samtools index fusion_genes.sorted.bam

# alignment done taking into account the paired-end information:

cat supporting_reads.fq | \
paste - - - - - - - - | \
awk '{print $1"\n"$2"\n"$3"\n"$4 > "r1.fq"; print $5"\n"$6"\n"$7"\n"$8 > "r2.fq"}'

bowtie2 \
--local \
-k 10 \
-x your_choice_of_genome_bowtie2_index \
-1 r1.fq \
-2 r2.fq \
-S fusion_genes.sam

samtools view -bS fusion_genes.sam | samtools sort - fusion_genes.sorted

samtools index fusion_genes.sorted.bam

• STAR aligner (where your_choice_of_genome_star_index should be built according to the STAR Manual)

# alignment done ignoring the paired-end information (i.e. like single reads):

STAR \
--genomeDir your_choice_of_genome_star_index \
--alignSJoverhangMin 9 \
--chimSegmentMin 17 \
--readFilesIn supporting_reads.fq \
--outFileNamePrefix .

samtools view -bS fusion_genes.sam | samtools sort - fusion_genes.sorted

samtools index fusion_genes.sorted.bam

# alignment done taking into account the paired-end information:

cat supporting_reads.fq | \
paste - - - - - - - - | \
awk '{print $1"\n"$2"\n"$3"\n"$4 > "r1.fq"; print $5"\n"$6"\n"$7"\n"$8 > "r2.fq"}'

STAR \
--genomeDir /your_choice_of_genome_star_index/ \
--alignSJoverhangMin 9 \
--chimSegmentMin 17 \
--readFilesIn r1.fq r2.fq\
--outFileNamePrefix .

samtools view -bS Aligned.out.sam | samtools sort - fusion_genes.sorted

samtools index fusion_genes.sorted.bam

• BLAT aligner (where your_choice_of_genome_blat_index should be built according to the BLAT’s examples)

# build the genome index using BLAT where the genome is given FASTA file genome.fa
faToTwoBit genome.fa genome.2bit -noMask


# align the supporting reads given by FusionCatcher (the FASTA 
# file for your fusion of interest can be found in ZIP files 
# generated as output by FusionCatcher, 
# e.g. EML4--ALK__42264951--29223528_reads.fa) using BLAT aligner
blat -stepSize=5 -repMatch=2253 -minScore=0 -minIdentity=0 genome.2bit supporting_reads.fa supporting_reads_mapped.psl 

# visualize the PSL file supporting_reads_mapped.pslin IGV or run psl2sam.pl to convert it into SAM format
psl2sam.pl supporting_reads_mapped.psl > supporting_reads_mapped.sam

Further, the files fusion_genes.sorted.bam and fusion_genes.sorted.bam.bai may be used with your favourite NGS visualizer!

6.3.4 - Chimera R/BioConductor package

For visualization of fusion genes found by FusionCatcher one may use also the R/BioConductor package Chimera, which supports FusionCatcher.

6.4 - Docker

Run FusionCatcher using Docker image, use the command::

docker run --rm fusioncatcher/docker fusioncatcher

In order to share a directory (for example: /data), use::

docker run --rm -v /data:/data fusioncatcher/docker fusioncatcher

6.5 - Examples

6.5.1 - Example 1

Here, is an example of how FusionCatcher can be used to search for fusion genes in human RNA-seq sample where: 1. any distance at chromosomal level between the candidate fusion genes is acceptable, and 1. the candidate fusion genes are allowed to be readthroughs (i.e. the genes forming a fusion gene maybe adjacent on the chromosome) 1. the candidate fusion genes are not allowed to be less the 1000 bp apart on the same strand 1. use two methods to find the fusion genes (i.e. use BOWTIE, BLAT, STAR, and BOWTIE2 aligners for mapping the reads and this allows to find the fusion genes even in the case that the annotation from Ensembl database is not entirely correct, like for example find a fusion junction even if it is in the middle of a exon or intron)

fusioncatcher \
-d /some/human/data/directory/ \
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/

6.5.2 - Example 2

Here, is an example of how FusionCatcher can be used to search for fusion genes in human RNA-seq sample where: 1. any distance at chromosomal level between the candidate fusion genes is acceptable, and 1. the candidate fusion genes are not allowed to be readthroughs (i.e. there is still at least one known gene situated one the same strand in between the genes which form the candidate fusion gene) 1. the candidate fusion genes are not allowed to be less the 1000 bp apart on the same strand 1. use only one method to find the fusion genes (i.e. use only BOWTIE aligner for mapping the reads and this allows to find the fusion genes only in the case that the annotation from Ensembl database is correct, like for example find a fusion junction only if it matches perfectly the known exon borders)

fusioncatcher \
-d /some/human/data/directory/ \
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/ \
--skip-readthroughs \
--skip-blat

7 - ALIGNERS

7.1 - Bowtie

By default, FusionCatcher its the Bowtie aligner for finding candidate fusion genes. This approach relies heavily on good is the annotation data for the given organism in the Ensembl database. If, for example, a gene is not annotated well and has several exons which are not annotated in the Ensembl database and if one of these exons is the one involved in the fusion point then this fusion gene will not be found by using only the Bowtie aligner. In order to find also the fusion genes where the the junction point is in the middle of exons or introns, *FusionCatcher* is using by default the BLAT, and STAR aligners in addition to Bowtie aligner. The command line options ‘--skip-blat’,‘--skip-star’, or ‘--skip-bowtie2’ should be used in order to specify what aligners should not be used. The command line option ‘--aligners’ specifies which aligners should be used by default. For example, ‘--aligners=blat,star,bowtie2’ forces FusionCatcher too use all aligners for finding fusion genes

7.2 - Bowtie and Blat

The use of Bowtie and Blat aligners is the default approach of FusionCatcher for finding fusion genes.

In order not to use this approach the command line option ‘--skip-blat’ should be added (or remove the string blat from line aligners from file fusioncatcher/etc/configuration.cfg), as following:

fusioncatcher \
-d /some/human/data/directory/ \ 
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/ \
--skip-blat

Please, read the license of Blat aligner before using this approach in order to see if you may use Blat! FusionCatcher will use Blat aligner when using this approach!

7.3 - Bowtie and STAR

The use of Bowtie and STAR aligners is the default approach of FusionCatcher for finding fusion genes.

In order not to use this approach the command line option ‘--skip-star’ should be added, as following:

fusioncatcher \
-d /some/human/data/directory/ \ 
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/ \
--skip-star

7.4 - Bowtie and Bowtie2

The use of Bowtie and Bowtie2 aligners is not the default approach of FusionCatcher for finding fusion genes.

In order not to use this approach the command line option ‘--skip-bowtie2’ should be added, as following:

fusioncatcher \
-d /some/human/data/directory/ \ 
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/ \
--skip-bowtie2

In order to use this approach the command line option ‘--aligners’ should contain the string ‘bowtie2’, like for example

fusioncatcher \
-d /some/human/data/directory/ \ 
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/ \
--aligners blat,star,bowtie2

7.5 - Bowtie2

The use of Bowtie2 aligner is not the default approach of FusionCatcher for finding fusion genes.

In order not to use this approach the command line option ‘--skip-bowtie2’ should be added, as following:

fusioncatcher \
-d /some/human/data/directory/ \ 
-i /some/input/directory/containing/fastq/files/ \
-o /some/output/directory/ \
--skip-bowtie2

9 - Methods

The main goal of FusionCatcher is to find somatic (and/or pathogenic) fusion genes in RNA-seq data.

FusionCatcher is doing its own quality filtering/trimming of reads. This is needed because most a very important factor for finding fusion genes in RNA-seq experiment is the length of RNA fragments. Ideally the RNA fragment size for finding fusion genes should be over 300 bp. Most of the RNA-seq experiments are designed for doing differentially expression analyses and not for finding fusion genes and therefore the RNA fragment size many times is less than 300bp and the trimming and quality filtering should be done in such a way that it does not decrease even more the RNA fragment size.

FusionCatcher is able to find fusion genes even in cases where the fusion junction is within known exon or within known intron (for example in the middle of an intron). The minimum condition for FusionCatcher to find a fusion gene is that both genes involved in the fusion are annotated in Ensembl database (even if their gene structure is not correct).

FusionCatcher is spending most of computational analysis on the most promising fusion genes candidate and tries as early as possible to filter out the candidate fusion genes which do not look promising, like for example: * candidate fusion gene is composed of a gene and its pseudogene, or * candidate fusion gene is composed of a gene and its paralog gene, or * candidate fusion gene is composed of a gene and a miRNA gene (but a gene which contains miRNA genes are not skipped), or * candidate fusion gene is composed of two genes which have a very sequence similarity (i.e. FusionCatcher is computing its homology score), or * candidate fusion gene is known to be found in samples from healthy persons (using the 16 organs RNA-seq data from the Illumina BodyMap2), or * candidate fusion gene is in one of the known databases of fusion genes found in healthy persons, i.e. ChimerDB2, CACG, and ConjoinG.

FusionCatcher is using by default three aligners for mapping the reads. The aligners are Bowtie, BLAT, and STAR. STAR is used here only and only for “splitting” the reads while aligning them.

10 - Comparisons to other tools

When performing comparisons where FusionCatcher is compared with other gene fusions finder we always recommend strongly to use the default/recommended parameters for FusionCatcher and also to use the raw FASTQ files which came directly from the Illumina sequencer.

The performance of FusionCatcher is decreased drastically, when using other parameters than the default/recommended ones! Especially do not change the defaults for: --5keep, --anchor-fusion, --reads-fusion, --pairs-fusion, --pairs-fusion2! The default parameters should work just fine for input reads which have the size range between 35 bp to 250 bp.

Also, when comparing the fusion genes found by FusionCatcher with fusion genes found by other tools one needs to keep in mind that FusionCatcher is a SOMATIC fusion gene finder and NOT a (general) fusion gene finder. This means that if a fusion gene is already known to exist in healthy individuals (from public literature or from our internal RNA-seq database of healthy sample) then that fusion gene will be skipped by FusionCatcher and it will not be reported at all! An example is the well known fusion gene TTTY15-USP9Y which is known to be found in healthy individuals (see here) and which FusionCatcher will skip it and will not report it on purpose because it is not a somatic fusion gene!

Also, when one is running FusionCatcher on some synthetic/simulated RNA-seq datasets which contain a set of random/ad-hoc fusion genes which are created randomly and without any biological support (for example, that fusion gene has never been reported in the literature to exist in a diseased patient), there most likely FusionCatcher will detect that these random/ad-hoc fusion genes are not fitting the already known biological knowledge (e.g. ad-hoc/random fusion gene might have been reported already to exist in healthy patients, or ad-hoc/random fusion is between a gene its paralog/homolog/pseudogene) and will skip them and will not report them even if it finds them. Therefore we strongly recommend not to run FusionCatcher on synthetic/simulated RNA-seq dataset which are known to contain fusion genes which are not somatic fusion genes. Also, we strongly recommend not to run FusionCatcher on downsampled input datasets, like for example, choosing randomly 30 million reads from an original datasets with 60 million reads. FusionCatcher has been specifically built for analyzing real input RNA-seq datasets which come directly from the sequencing machine.

11 - License

FusionCatcher’s code is released under GNU GPL version 3 license. FusionCatcher is using third-party tools and databases. The user is responsible to obtain licenses for the third-party tools and databases which are used by FusionCatcher.

Most (but not all) of the third-party tools and databases used by FusionCatcher are (i) free to use, or (ii) are released under GPL/MIT-type licenses. The most notable exception here of which we are aware is BLAT’s aligner license, which requires one to buy a license when BLAT is used in commercial environment (please, see for more here). In case that one does not wish to use BLAT aligner then it is still possible to use FusionCatcher for finding fusion genes, by telling FusionCatcher not to use BLAT aligner but instead to use the BOWTIE2 aligner (BLAT is used by default and BOWTIE2 is not used by default), as following:

/apps/fusioncatcher/bin/fusioncatcher \
--aligners star,bowtie2

12 - Citing

If you use FusionCatcher, please cite:

D. Nicorici, M. Satalan, H. Edgren, S. Kangaspeska, A. Murumagi, O. Kallioniemi, S. Virtanen, O. Kilkku, FusionCatcher – a tool for finding somatic fusion genes in paired-end RNA-sequencing data, bioRxiv, Nov. 2014, DOI:10.1101/011650

13 - Reporting Bugs

Please, when reporting bugs include also the following files: * “fusioncatcher.log” (this contains just a list of the commands executed by FusionCatcher), and * “info.txt” (this contains info: regarding the version of FusionCatcher and tools used, statistics about input FASTQ files, counts of found reads, etc.) which were generated by FusionCatcher during the run.

NOTE: Giving only step number where the error has appeared is not enough because the step numbers depend on the input type (e.g. raw FASTQ file, ZIP compressed FASTQ file, SRA file, etc.) and command line options used to run FusionCatcher (e.g. some command line option skip some steps). The step numbers are used by FusionCatcher for being able to re-start from the last step which was executed successfully last time in case that last time the run ended prematurely due to reasons which didn’t depend on FusionCatcher (e.g. server crashed).

NOTES

• The performance of FusionCatcher is decreased drastically, when using other parameters than the default/recommended ones! Especially do not change the defaults for: –5keep, –anchor-fusion, –reads-fusion, –pairs-fusion, –pairs-fusion2! The default parameters should work just fine for input reads which have the size range between 35 bp to 250 bp.
• • The performance of FusionCatcher is decreased drastically, when running FusionCatcher on a subset of the reads. It is not recommended to run FusionCatcher on 20 million paired-reads sampled from a sample. All the reads from the sample shall be given as input to FusionCatcher
• fusioncatcher-build takes several hours to run and it depends on the local internet connection speed. It needs to be run only once!
• FusionCatcher can be run many times using the same data produced by the fusioncatcher-build;
• Ensembl version 98 was found to work fine with FusionCatcher as November 2019;
• FusionCatcher and fusioncatcher-build restart automatically from the point where have been interrupted at the previous run.
• FusionCatcher by default is focusing on finding fusion genes specific to diseased/tumor/cancer samples. That means that FusionCatcher will skip the fusion genes which are already known to exist in healthy samples. If one wishes to find fusion genes in healthy samples then we suggest other fusion finders to be used.
• FusionCatcher is able to find fusion genes also without using BLAT aligner but in this case we recommend to user BOWTIE2 aligner (which is not used by default) also in order to compensate!