FusionCatcher

Overview

FusionCatcher is a well-known tool that searches for somatic novel/known fusion genes, translocations, and/or chimeras in RNA-seq data. While FusionCatcher itself is not part of Illumina Connected Annotations, we have included a subset of their genomic databases in Illumina Connected Annotations.

Publication

Daniel Nicorici, Mihaela Şatalan, Henrik Edgren, Sara Kangaspeska, Astrid Murumägi, Olli Kallioniemi, Sami Virtanen, Olavi Kilkku. (2014) FusionCatcher – a tool for finding somatic fusion genes in paired-end RNA-sequencing data. bioRxiv 011650

Supported Data Sources

Oncogenes

The following data sources are aggregated and used to populate the isOncogene field in the gene JSON object:

Description	Reference	Data	FusionCatcher filename
Bushman		bushmanlab.org	cancer_genes.txt
ONGENE	JGG	bioinfo-minzhao.org	oncogenes_more.txt
UniProt tumor genes	NAR	uniprot.org	tumor_genes.txt

Germline

Illumina Connected Annotations label	Reference	Data	FusionCatcher filename
1000 Genomes Project	PLOS ONE		1000genomes.txt
Healthy (strong support)			banned.txt
Illumina Body Map 2.0		EBI	bodymap2.txt
CACG	Genomics		cacg.txt
ConjoinG	PLOS ONE		conjoing.txt
Healthy prefrontal cortex	BMC Medical Genomics	NCBI GEO	cortex.txt
Duplicated Genes Database	PLOS ONE	genouest.org	dgd.txt
GTEx healthy tissues		gtexportal.org	gtex.txt
Healthy			healthy.txt
Human Protein Atlas	MCP	EBI	hpa.txt
Babiceanu non-cancer tissues	NAR	NAR	non-cancer_tissues.txt
non-tumor cell lines			non-tumor_cells.txt
TumorFusions normal	NAR	NAR	tcga-normal.txt

Somatic

Illumina Connected Annotations label	Reference	Data	FusionCatcher filename
Alaei-Mahabadi 18 cancers	PNAS		18cancers.txt
DepMap CCLE		depmap.org	ccle.txt
CCLE Klijn	Nature Biotechnology	Nature Biotechnology	ccle2.txt
CCLE Vellichirammal	Molecular Therapy Nucleic Acids		ccle3.txt
Cancer Genome Project		COSMIC	cgp.txt
ChimerKB 4.0	NAR	kobic.re.kr	chimerdb4kb.txt
ChimerPub 4.0	NAR	kobic.re.kr	chimerdb4pub.txt
ChimerSeq 4.0	NAR	kobic.re.kr	chimerdb4seq.txt
COSMIC	NAR	COSMIC	cosmic.txt
Bao gliomas	Genome Research		gliomas.txt
Known			known.txt
Mitelman DB	ISB-CGC	Google Cloud	mitelman.txt
TCGA oesophageal carcinomas	Nature		oesophagus.txt
Bailey pancreatic cancers	Nature	Nature	pancreases.txt
PCAWG	Cell	ICGC	pcawg.txt
Robinson prostate cancers	Cell	Cell	prostate_cancer.txt
TCGA		cancer.gov	tcga.txt
TumorFusions tumor	NAR	NAR	tcga-cancer.txt
TCGA Gao	Cell	Cell	tcga2.txt
TCGA Vellichirammal	Molecular Therapy Nucleic Acids		tcga3.txt
TICdb	BMC Genomics	unav.edu	ticdb.txt

Gene Pair TSV File

Most of the data files in FusionCatcher are two-column TSV files containing the Ensembl gene IDs that are paired together.

Example

Here are the first few lines of the 1000genomes.txt file:

ENSG00000006210 ENSG00000102962
ENSG00000006652 ENSG00000181016
ENSG00000014138 ENSG00000149798
ENSG00000026297 ENSG00000071242
ENSG00000035499 ENSG00000155959
ENSG00000055211 ENSG00000131013
ENSG00000055332 ENSG00000179915
ENSG00000062485 ENSG00000257727
ENSG00000065978 ENSG00000166501
ENSG00000066044 ENSG00000104980

Parsing

In Illumina Connected Annotations, we will only import a gene pair if both Ensembl gene IDs are recognized from either our GRCh37 or GRCh38 cache files.

Gene TSV File

Some of the data files are single-column files containing Ensembl gene IDs. This is commonly used in the data files representing oncogene data sources.

Example

Here are the first few lines of the oncogenes_more.txt file:

ENSG00000000938
ENSG00000003402
ENSG00000005469
ENSG00000005884
ENSG00000006128
ENSG00000006453
ENSG00000006468
ENSG00000007350
ENSG00000008294
ENSG00000008952

Parsing

Known Issues

Known Issues

FusionCatcher also uses creates custom Ensembl genes (e.g. ENSG09000000002) to handle missing Ensembl genes. Illumina Connected Annotations will ignore these entries since we only include the gene IDs that are currently recognized by Illumina Connected Annotations.

I suspect that these were originally RefSeq genes and if so, we can support those directly in Illumina Connected Annotations in the future.

Download URL

https://sourceforge.net/projects/fusioncatcher/files/data

JSON Output

   "fusionCatcher":[
      {
         "genes":{
            "first":{
               "hgnc":"ETV6",
               "isOncogene":true
            },
            "second":{
               "hgnc":"RUNX1"
            },
            "isParalogPair":true,
            "isPseudogenePair":true,
            "isReadthrough":true
         },
         "germlineSources":[
            "1000 Genomes Project"
         ],
         "somaticSources":[
            "COSMIC",
            "TCGA oesophageal carcinomas"
         ]
      }
   ]

Field	Type	Notes
genes	genes object	5' gene & 3' gene
germlineSources	string array	matches in known germline data sources
somaticSources	string array	matches in known somatic data sources

genes

Field	Type	Notes
first	gene object	5' gene
second	gene object	3' gene
isParalogPair	bool	true when both genes are paralogs for each other
isPseudogenePair	bool	true when both genes are pseudogenes for each other
isReadthrough	bool	true when this fusion gene is a readthrough event (both are on the same strand and there are no genes between them)

gene

Field	Type	Notes
hgnc	string	gene symbol. e.g. MSH6
isOncogene	bool	true when this gene is an oncogene