CancerGenomeProject

COSMIC v46 Release

2010-03-08T09:42:18Z

The second full-genome resequencing study from the CGP at the Sanger Institute, UK is now available, together with the curation of Parsons et al (2008), a systematic candidate gene screen of Glioblastomas. In addition, the published literature has been fully curated for fusion mutations between seven new gene pairs.

Full Genome resequencing of NCI-H209

The recent Pleasance et al (2010) publication "A small-cell lung cancer genome with complex signatures of tobacco exposure" (Nature 463, 184-190) is now available within COSMIC; please click here.

Systematic Screen Curation

The largest published candidate gene screen of Glioblastomas Parsons et al (2008), is now curated in COSMIC; please click here:

An integrated genomic analysis of human glioblastoma multiforme. Parsons DW, Jones S, Zhang X, Lin JC, Leary RJ, Angenendt P, Mankoo P, Carter H, Siu IM, Gallia GL, Olivi A, McLendon R, Rasheed BA, Keir S, Nikolskaya T, Nikolsky Y, Busam DA, Tekleab H, Diaz LA, Hartigan J, Smith DR, Strausberg RL, Marie SK, Shinjo SM, Yan H, Riggins GJ, Bigner DD, Karchin R, Papadopoulos N, Parmigiani G, Vogelstein B, Velculescu VE, Kinzler KW Science. 2008;321;1807-12. PMID: 18772396 DOI: 10.1126/science.1164382

Statistics

Samples	105
Mutations	2449

Fusion mutations between 7 new gene pairs have been curated from the literature for this release.

FUS-ERG , FUS-FEV , FUS-ATF1 Both FUS-ERG and FUS-FEV fusions have been identified as alternatives to EWSR1-ETS transcription factor fusions in Ewing's sarcoma, and FUS-ERG also occurs in t (16,21) myeloid leukaemia as well as in these solid tumours. FUS-ATF1 is found in angiomatoid fibrous histiocytoma, where the fusion of the N-terminus of FUS and the DNA binding domain of ATF1 is similar to the EWSR1-ATF1 fusion found in clear cell sarcoma.

SS18-SSX1 This fusion is characteristic for synovial sarcoma along with SS18-SSX2 and more rarely, SS18-SSX4 fusions. Through its N-terminal SNH domain SS18 protein is involved in the remodelling of chromatin structures and functions as a transcriptional activator whereas SSX proteins have 2 putative transcription-repressor domains, one of which, an SSXRD domain in the C-terminal region, is preserved in the fusion protein.

SRGAP3-RAF1 This oncogenic fusion has been identified in paediatric pilocytic astrocytoma as an alternative to the previously described KIAA1549-BRAF fusion. It also activates the ERK/MAPK pathway; the auto-inhibitory domain of RAF1 being replaced by SRGAP3.

COL1A1-PDGFB This recurrent fusion characterizes dermatofibroma protuberans and its juvenile form, giant cell fibroblastoma. The fusion consistently deletes exon 1 of PDGFB releasing this growth factor from its normal regulation. The breakpoints in COL1A1, which encodes an extracellular matrix protein, occur in various exons in the alpha-helical domain.

JAZF1-SUZ12 A fusion involving these two genes is common but not universal in endometrial stromal sarcomas, occurring less frequently in high-grade tumours. The genes encode novel proteins with zinc finger motifs and these are retained in the fusion.

The following curated genes have been updated in this release

ABL1, ACVR1B, AKT1, ALK, APC, ASXL1, ATM, BRAF, BRCA1, BRCA2, CBL, CDC73, CDH1, CDKN2A, CEBPA, CSF1R, CTNNA1, CTNNB1, CYLD, EGFR, EML4, ERBB2, ERG, FAM123B, FBXW7, FGFR1, FGFR2, FGFR3, FLT3, FOXL2, GATA1, GNAQ, GNAS, HNF1A, HRAS, IDH1, IDH2, JAK2, JAK3, KIT, KRAS, MAP2K4, MEN1, MET, MLH1, MPL, MSH2, MSH6, NF1, NF2, NOTCH1, NOTCH2, NPM1, NRAS, PDGFRA, PHOX2B, PIK3CA, PRKAR1A, PTCH1, PTEN, PTPN11, RB1, RET, RUNX1, SMAD4, SMARCA4, SMARCB1, SMO, SOCS1, SRC, STK11, SUFU, TNFAIP3, TSHR, VHL, WT1

COSMIC v46 Total Statistics

Experiments	2077858
Tumours	449676
Samples	451972
Mutant Samples	108773
Mutations	112256
Unique Mutations	19239
Papers curated	8911
Genes	18478
Fusions	4657
Structural Variants	2307

COSMIC v45 Release

2010-01-21T17:59:00Z

The first full-genome resequencing study is now available, together with the genome-wide rearrangement screens of 24 breast tumours. In addition, five new cancer genes have been curated from the literature.

To make the data easier to investigate in depth, the website has been upgraded with new specialisation features, together with new views on mutation spectrum and distribution. Finally, we are introducing a new COSMIC Biomart, where all COSMIC's information will be available in this industry-standard data mining tool.

Full Genome resequencing of COLO-829

The recent Pleasance et al (2010) publication "A comprehensive catalogue of somatic mutations from a human cancer genome" (Nature 463, 191-196) is now available within COSMIC; please click here.

Whole-genome rearrangement screen of 24 Breast tumour samples:

Also, the CGP Stephens et al (2009) paper "Complex landscapes of somatic rearrangement in human breast cancer genomes" (Nature 462, 1005-1010) is now available in COSMIC; please click here . A paired-end genome-wide Illumina sequencing strategy revealed numerous rearrangements in very diverse patterns between the samples examined.

New genes curated from the scientific literature

GNAQ is the alpha subunit of one of the heterotrimeric GTP-binding proteins that mediate stimulation of protein kinase C signalling. Mutations in GNAQ, occurring at codon 209 in the catalytic domain, have been found as common and early mutational events in uveal melanomas.

TNFAIP3 is a negative regulator of the NF-kappa B pathway functioning through the removal of activating Lys63-linked ubiquitins and the Lys48-linked ubiquitination of receptor-interacting proteins. TNFAIP3 has been shown to be a genetic target in B-lineage lymphomas such as mucosa-associated lymphoma and Hodgkin's lymphoma of nodular sclerosing histology.

CBL encodes a protein with multiadaptor function and E3 ubiquitin ligase activity that targets a variety of tyrosine kinases for degradation. Mutations in CBL have been identified in myeloid malignancies, occurring in the critical linker and ring finger domains of the protein.

JAK3 is a member of the non-receptor tyrosine kinase family which includes JAK2. Rare but significant JAK3 activating mutations located in the JH2 (pseudokinase) and JH6 (receptor binding) domains have been found in Down syndrome and Non-DS acute megakaryoblastic leukaemia (AML-M7). Mutations have also been found in various myeloproliferative neoplasms, lymphomas and carcinomas.

NOTCH2 is a Type 1 transmembrane protein with an extracellular domain consisting of multiple epidermal growth factor-like (EGF) repeats, and an intracellular domain consisting of multiple different domain types. The Notch2 receptor and its 5 ligands, which include Jagged1, Jagged2, and Delta-like 1, 3 and 4, send signals that are important for development before birth. After birth,Notch2 signaling is involved in tissue repair. Mutations in the NOTCH2 gene have been identified in a small percentage of people with Alagille syndrome and malformations in the kidneys, especially in filtering structures. NOTCH2 is also preferentially expressed in mature B cells,is essential for marginal zone B-cell generation, and mutations are evident in a subset of individuals with diffuse large B-cell lymphomas.

Web site enhancements

The main histogram page of the COSMIC website had been improved to provide better ways of selecting and viewing subsets of data. In the navigation bar on the left side, new options are now available to redraw the histogram and associated tables based on four parameters: mutation type (eg deletion, nonsense substitutions, etc), sample source (cultured or tissue sample), somatic status (confirmed somatic or unknown) and systematic screen (genome-wide screen). In addition to redrawing the histogram and tables, a new "Distribution" button displays pie charts of relevant information about the data selected.

The sample summary page has also been upgraded, with every CGP sample (examined through numerous genes) receiving a mutation spectrum diagram. This comprises a histogram showing the relative frequencies of each substitution type, together with a count of insertion/deletion mutations. This is highly useful when looking for mutation signatures which may show characteristsics of, for instance, tobacco or UV light exposure.

Biomart

The new COSMIC biomart is now available, please click here. This system allows much more specialised selection of data in COSMIC and is very useful for data mining. In addition, it can be directly linked to Ensembl for federilsed querying across both databases.

The following curated genes have been updated in this release

JAK2, JAK3, MAP2K4, GNAS, MPL, SOCS1, WT1, CYLD, FBXW7, MEN1, NF1, RUNX1, ASXL1, NOTCH2, IDH1, IDH2, APC, CDH1, VHL, GNAQ, BRAF, HRAS, CEBPA, CTNNB1, FLT3, KIT, PDGFRA, PTEN, RB1, RET, SMARCB1, AKT1, EGFR, ERBB2, CDKN2A, CBL, GATA1, NPM1, PTPN11, NRAS, FGFR3, BRCA1, MSH6, PRKAR1A, KRAS, PIK3CA, MET, TNFAIP3

COSMIC v45 Total Statistics

Experiments	1654274
Tumours	434364
Samples	436577
Mutant Samples	101860
Mutations	105171
Unique Mutations	16788
Papers curated	8624
Genes	13634
Fusions	3635
Structural Variants	2249

COSMIC v44 Release

2009-11-04T12:39:17Z

This release of COSMIC includes 4 new curated genes, 8 new curated fusion pairs and the TCGA systematic screen publication of 91 Glioblastoma tumour samples. In addition, a new CGP study is available (Adenoid cystic carcinoma) together with substantial updates to existing data.

New curated genes

IDH2 encodes a mitochondrial NADP(+)-dependent isocitrate dehydrogenase which catalyzes oxidative decarboxylation of isocitrate to alpha-ketoglutarate. It is now implicated in the pathogenesis of malignant gliomas and some secondary glioblastomas lacking IDH1 mutations have IDH2 mutations at the analogous amino acid (R172).

AKT1 encodes a serine-threonine protein kinase which is activated by phosphorylated phosphoinositides and is a central mediator of the PI3kinase signalling pathway. A common mutation (E17K) has been identified in the pleckstrin homology domain in cancers of the colon, breast, lung and ovary.

ASXL1 belongs to a family of proteins regulating chromatin remodelling. Originally implicated via aCGH on MDS/AML samples, mutations are mainly frameshift mutations, the predicted truncated proteins lack the PHD finger domain potentially compromising the function of the associated chromatin modifiers.

FOXL2, forkhead box L2 is a winged helix/forkhead transcription factor gene, encoding a nuclear protein that is specifically expressed in eyelids and in fetal and adult ovarian follicular cells. Germline mutations in FOXL2 are responsible for BPES - blepharophimosis ptosis epicanthus inversus syndrome - an autosomal dominant disorder consisting of eyelid abnormalities (only, in Type II) and ovarian failure (Type I). Somatic mutations have recently been described in ovarian granulosa cell tumours.

New curated gene fusion pairs:

The following gene fusions have been curated from the scientific literature:
EML4 / ALK
MSN / ALK
NPM1 / ALK
CLTC / ALK
SEC31A / ALK
RANBP2 / ALK
SS18 / SSX2
SS18 / SSX4

Systematic screen curation:

Comprehensive genomic characterization defines human glioblastoma genes and core pathways.The first systematic screen of the Cancer Genome Atlas Research Network (PMID 18772890) is now curated in COSMIC .

Comprehensive genomic characterization defines human glioblastoma genes and core pathways.
Cancer Genome Atlas Research Network
Nature. 2008;455;1061-8. PMID: 18772890 DOI: 10.1038/nature07385

Statistics

Glioblastoma Samples	91
Genes	599
Sequencing Experiments	54509
Mutations	662

New CGP resequencing study: Adenoid Cystic Carcinoma Candidate Gene Screen

Adenoid cystic carcinoma is a slow growing tumour of the secretory glands, arising most commonly in the salivary glands but also occurring in other parts of the body. As part of an ongoing research effort funded by the Adenoid Cystic Carcinoma Research Fund (www.accrf.org), 400 candidate gene (including genes implicated in cancer, cell signaling and growth control) were sequenced for small point mutations. This work was carried out on 25 samples (provided by ACCRF collaborative research group member Dr. Adel El-Naggar) utilising an approach of PCR product generation for the entire set of PCR amplimers followed by individual concatentation of all amplimers for each tumour and matching normal DNA sample, then sequencing this material utilising next generation sequencing. In total 8 somatic point mutations were identified in 8 genes. No highly prevalent point mutation was identified in this set of genes.

These curated genes have been updated this release

KRAS, PIK3CA, FGFR2, MET, ABL1, FGFR1, JAK2, MAP2K4, GNAS, EML4, FOXL2, PTCH1, MPL, SOCS1, HNF1A, WT1, NF2, CYLD, FBXW7, MEN1, NF1, RUNX1, IDH1, IDH2, ASXL1, FAM123B, APC, CDH1, SMAD4, VHL, BRAF, HRAS, CEBPA, CSF1R, CTNNB1, FLT3, KIT, PDGFRA, PTEN, RB1, RET, SMARCB1, SUFU, ACVR1B, AKT1, ALK, ATM, EGFR, ERBB2, SRC, STK11, CDKN2A, GATA1, SMO, NOTCH1, NPM1, PTPN11, NRAS, FGFR3, BRCA1, BRCA2, MLH1, MSH2, MSH6, PRKAR1A

COSMIC v44 Total Statistics

Experiments	1631186
Tumours	419018
Samples	421193
Mutant Samples	97932
Mutations	101138
Unique Mutations	16072
Papers curated	8336
Genes	13501
Fusions	3521
Structural Variants	40

COSMIC v43 Release

2009-08-26T16:55:18Z

The COSMIC curation systems have been extended to encompass the entry of large-scale systematic screen papers. For this release, we have entered the first such paper, the Sjoblom et al (2006) screen of human breast and colorectal cancers. This release also contains two new genes successfully curated from the scientific literature (IDH1, SMARCA4) and the finalisation of two of the Cancer Genome Project's current resequencing studies.

Systematic Screen Papers Curated in COSMIC

For this release of COSMIC we have entered the Sjoblom et al (2006) systematic screen paper of human breast and colorectal cancers. An additional 8,648 genes have been added to COSMIC along with the 1,672 mutations from the paper. The COSMIC reference overview page for this publication is available here.

The consensus coding sequences of human breast and colorectal cancers. Sjoblom T, Jones S, Wood LD, Parsons DW, Lin J, Barber TD, Mandelker D, Leary RJ, Ptak J, Silliman N, Szabo S, Buckhaults P, Farrell C, Meeh P, Markowitz SD, Willis J, Dawson D, Willson JK, Gazdar AF, Hartigan J, Wu L, Liu C, Parmigiani G, Park BH, Bachman KE, Papadopoulos N, Vogelstein B, Kinzler KW, Velculescu VE. Science. 2006 Oct 13;314(5797):268-74. Epub 2006 Sep 7. PMID: 16959974

CGP resequencing studies completed

The resequencing of candidate genes in Pilot and Renal tumour sets has now been completed. The finalised studies examined 2978 samples through 4766 genes, discovering a total of 5437 mutations. All of these can be found in COSMIC's CGP Resequencing Studies Site.

New curated genes

IDH1 is a catalytic enzyme causing NADP+ dependent oxidative decarboxylation of isocitric acid. It plays an important role in the control of glucose-stimulated insulin secretion and the cholesterol and fatty acid biosynthetic pathways. Originally implicated in human cancer in genome-wide sequencing scans, when mutated it is an indicator for the longer survival of these patients.

SMARCA4, is a scaffold protein, forming a functional part of the SWI/SNF complex involved in the control of transcription.

These curated genes have been updated this release

FBXW7, MEN1, NF1, BRAF, HRAS, CSF1R, CTNNB1, FLT3, KIT, PDGFRA, PTEN, RET, SMARCB1, SUFU, ACVR1B, ATM, EGFR, ERBB2, SRC, CDKN2A, FAM123B, GATA1, SMO, NOTCH1, NPM1, PTPN11, NRAS, FGFR3, BRCA1, BRCA2, APC, CDH1, SMAD4, VHL, TSHR, MLH1, MSH2, MSH6, SMARCA4, RUNX1, PHOX2B, GNAS, KRAS, PIK3CA, FGFR2, FGFR1, IDH1, JAK2, JAK3, MAP2K4, TET2, PRKAR1A, CDC73, PTCH1, MPL, CTNNA1, SOCS1, HNF1A, WT1, ERG, NF2

COSMIC v43 Total Statistics

Experiments	1506545
Tumours	366477
Samples	368592
Mutant Samples	85749
Mutations	88727
Unique Mutations	14971
Papers curated	7797
Genes	13423
Fusions	2770
Structural Variants	40

COSMIC v42 Release

2009-05-28T13:59:06Z

For this release of COSMIC two known cancer genes (GNAS and ALK) and 3 gene fusions (FCHSD1 / BRAF, KIAA1549 / BRAF, EWSR1 / NR4A3) have been successfully curated from the scientific literature. The Cancer Cell Line Project has also been updated with the addition of 80 mutations.

Cancer Cell Line Project Update

The Cancer Cell Line Data has been updated with the addition of 80 mutations. The project has also published a further set of variants identified by the screen which have been classified as Tentatively Oncogenic Variant (TOV) or Unknown Variant (UV). These variants are currently available from our website as an excel file.

Curation of known cancer genes ALK and GNAS

Two further cancer genes have been curated with the addition of 95 mutations for ALK and 235 mutations for GNAS.

Curation of gene fusions

The following gene fusions have been curated from the scientific literature:
FCHSD1 / BRAF
KIAA1549 / BRAF
EWSR1 / NR4A3

Genes updated: KRAS, PIK3CA, ABL1, FGFR1, JAK2, BRAF, HRAS, CEBPA, CSF1R, CTNNB1, KIT, PDGFRA, PTEN, RB1, SUFU, ERBB2, SRC, STK11, CDKN2A, GATA1, SMO, NPM1, PTPN11, NRAS, BRCA2, MLH1, MSH2, MSH6, APC, CDH1, SMAD4, MET, EGFR, FLT3, PTCH1, MPL, WT1, CYLD, FBXW7, NF1, ALK, FGFR3, RET, NOTCH1, NF2, GNAS

COSMIC v42 Total Statistics

Experiments	1111579
Tumours	339481
Samples	341522
Mutant Samples	76132
Mutations	78933
Unique Mutations	12905
Papers curated	7386
Genes	4775
Fusions	2424
Structural Variants	40

COSMIC v41 release

2009-03-04T15:59:41Z

This release of COSMIC comprises an update of published data in which 44 genes have been updated with the addition of 22516 samples and a further 7387 mutations.

Gene Update

STK11, CDKN2A, GATA1, SMO, NPM1, PTPN11, NRAS, BRCA2, MSH2, KRAS, PIK3CA, JAK2, MAP2K4, BRAF, HRAS, CEBPA, CTNNB1, KIT, PDGFRA, PTEN, RB1, ATM, ERBB2, FBXW7, NF1, FAM123B, APC, CDH1, VHL, MET, EGFR, FLT3, PTCH1, MPL, SOCS1, HNF1A, WT1, CYLD, FGFR3, RET, RUNX1, TSHR, PHOX2B, NOTCH1.

COSMIC v41 Total Statistics

Experiments	1078748
Tumours	313780
Samples	315778
Mutant Samples	70086
Mutations	72718
Unique Mutations	12349
Papers curated	6876
Genes	4773
Fusions	2266
Structural Variants	40

COSMIC release 39, Annotating Cancer Genomes

2008-10-15T14:59:34Z

For this release of COSMIC the database and web interfaces have been upgraded to handle Next Generation Sequencing Data. This is part of ongoing work to allow COSMIC to handle the increased volumes and complexity of somatic data that is anticipated from Next Generation Sequencers. In particular, for this release we have concentrated on adapting COSMIC to handle large-scale structural variants (including translocations, large insertions/deletions, inversions, and duplications).

The structural variants from the Campbell et al. 2008 paper, which comprehensively characterizes 2 lung cancer cell lines, have been entered into COSMIC (click here for study overview). Sample Summary pages are available for both cancer cell lines (NCI-H2171 and NCI-H1770).

Circular plots (Circos plots developed by Martin Krzywinski) have been added to the sample overview page which gives a clear overview of all the structural variants along with copy number changes and COSMIC point mutations for a particular sample (Figure 1). More detailed views of complex rearrangements are available on the mutation details page.

Figure 1. Circos Plot showing structural variants in relation to copy number and COSMIC Point Mutations.

Tabular views and exports are also available for these data (Figure 2). Due to the complexity of these rearrangements, where possible, a short description term of the variant is given (e.g. deletion, tandem duplication translocation). The variant is also fully described using HGVS mutation nomenclature. For example chr11:g.36585230_76606619del, where chr11: denotes the chromosome involved, g. for genomic coordinates, 36585230 for the deletion start point, 76606619 for deletion end point and del indicates a deletion event.

Figure 2. Summary Structural Variants Table

Bioinformatics Primer on COSMIC published

NCI/Nature Pathway Interaction Database Primer on COSMIC published and is available from here.

Update of the Cancer Gene Census

The Cancer Gene Census was updated on 11th August 2008. The Census now contains information of 379 genes of which 343 harbour somatic alterations and 70 germline.

COSMIC v39 General Statistics

Experiments	1035943
Tumours	281307
Samples	282777
Mutant Samples	60007
Mutations	62352
Unique Mutations	11642
Papers curated	6168
Genes	4773
Fusions	2266
Structural Variants	40

COSMIC release 38

2008-07-03T16:59:20Z

For this release of COSMIC we have concentrated our efforts on significantly updating the following genes: BRAF, HRAS, CTNNB1, KIT, PDGFRA, PTEN, RB1, ERBB2, MAP2K4, CDKN2A, GATA1, SMO, NPM1, NRAS, MLH1, MSH2, KRAS, PIK3CA, JAK2, APC, SMAD4, EGFR, FLT3, PTCH1, MPL, HNF1A, FBXW7, NF1, FGFR3, RET, NF2, NOTCH1.

External links

In collaboration with the Human Gene Nomenclature committee (HGNC) and the Atlas of Genetics and Cytogenetics in Oncology and Haematology (Atlas Genetics Oncology), links are now available from COSMIC's gene summary page to further information at these resources.

A Current Protocol for COSMIC

An article describing COSMIC, its contents and usage, has been published in Current Protocols in Human Genetics, unit 10.11. Describing in detail how the website and exported datasheets may be used and interpreted, this is available at the Wiley Interscience website.

COSMIC v38 total statistics

Experiments	1019304
Tumours	268938
Samples	270095
Mutant Samples	56918
Mutations	59187
Unique Mutations	11400
Papers curated	5902
Genes	4773
Fusions	2266

COSMIC release 37

2008-05-07T14:59:30Z

This months release extends our complete curation of oncogenic EWSR1 fusion partners, together with two new curated genes, PHOX2B & PRKAR1A. CGP's resequencing studies and cell line projects are also significantly updated, each receiving over 100 new mutations. In total, over 1200 new mutations have been added to COSMIC this release.

Curated genes

PHOX2B This gene encodes a highly conserved homeobox transcription factor known to cause congenital central hypoventilation syndrome with associated neuroblastoma.

PRKAR1A This is a regulatory subunit of the cAMP dependent protein kinase holoenzyme. An apparent tumour suppressor gene, it has also been observed to be oncogenic in fusions with RET and RARA.

Gene	Unique Samples	Samples	Experiments	Mutants	Papers	Mutations	Unique Mutations
PHOX2B	410	410	411	6	4	6	5
PRKAR1A	232	232	233	7	5	7	7

Curated EWSR1 fusions

EWSR1/ETV4; EWSR1/FEV; EWSR1/PATZ1; EWSR1/PBX1; EWSR1/POU5F1; EWSR1/ZNF384

EWSR1 has been observed in oncogenic gene fusions with over 15 partners. This month we release our curation of the literature describing its fusion with a further six partners, bringing the total to 14.

Genes	Unique Breakpoints	Mutations	Unique Fusions	Papers	Mutant Samples
EWSR1 / ETV4	3	6	4	2	3
EWSR1 / FEV	5	10	4	4	5
EWSR1 / PATZ1	1	2	2	1	1
EWSR1 / PBX1	1	3	3	1	1
EWSR1 / POU5F1	5	10	4	2	5
EWSR1 / ZNF384	2	4	4	1	2

The following curated genes have received significant updates: BRAF, HRAS, KIT, PTEN, RB1, SMARCB1, ERBB2, STK11, CDKN2A, PTPN11, NRAS, BRCA2, MLH1, MSH2, KRAS, PIK3CA, JAK2, APC, VHL, MSH6, MET, EGFR, MPL, FBXW7, PRKAR1A, RET, RUNX1, NOTCH1, NF2, PHOX2B.

COSMIC v37 total statistics

Experiments	1006553
Tumours	258584
Samples	259684
Mutant Samples	53569
Mutations	55779
Unique Mutations	11207
Papers curated	5706
Genes	4773
Fusions	2249

COSMIC release 36

2008-03-05T14:59:26Z

The March 2008 release of COSMIC contains full curation of the TSHR gene together with a further 6 EWSR1 gene fusion pairs.

Curated genes

TSHR - Thyroid stimulating hormone receptor is a 7-TM cell surface receptor expressed in follicular thyroid cells. Upon binding of its ligand, thyrotropin, a signalling cascade is commenced resulting in a range of transcriptional alterations. Somatic mutations in this gene have been described in thyroid adenomas and carcinomas.

Gene	Samples	Experiments	Mutants	Papers	Mutations	Unique Mutations
TSHR	665	669	210	36	210	61

Curated Fusions

EWSR1/ATF1 ; EWSR1/CREB1 ; EWSR1/DDIT3 ; EWSR1/ETV1 ; EWSR1/SP3 ; EWSR1/WT1
EWSR1 is fused to multiple partner genes via recurrent chromosomal translocation in, primarily, Ewing sarcoma. We are currently curating the complete mutation data for this gene, which has so far been fused with over 10 partners; we have released our curation of EWSR1 with ERG & FLI1, we now release the data for six more gene partners.

Genes	Mutant Samples	Mutations	Unique fusions	Papers
EWSR1 / ATF1	72	175	16	17
EWSR1 / CREB1	24	36	5	3
EWSR1 / DDIT3	11	22	7	6
EWSR1 / ETV1	4	7	3	3
EWSR1 / SP3	1	3	3	1
EWSR1 / WT1	102	198	22	28

The following curated genes have received significant updates:
BRAF, BRCA1, BRCA2, CDH1, CDKN2A, CEBPA, EGFR, ERBB2, FLT3, HRAS, KRAS, MLH1, MSH2, MSH6, NF2, NRAS, PDGFRA, PTEN, SMARCB1, STK11, TSHR, VHL

COSMIC v36 total statistics

Experiments	1000842
Tumours	254672
Samples	255767
Mutant Samples	52343
Mutations	54528
Unique Mutations	10995
Papers curated	5614
Genes	4772
Fusions	2174

COSMIC release 35

2008-01-16T14:59:00Z

This release of COSMIC contains the new curation of four new tumour suppressor genes, and further curation of EWSR1/FLI1 gene fusions in Ewing's sarcoma. We also announce a significant upgrade to the CGP Trace Archive, which is now updated daily with our latest sequencing results.

Literature Curation

MLH1

MLH1 is a tumour suppressor gene, involved in mismatch repair. The encoded protein is a subunit of the large 'BRCA1-associated genome surveillance complex' (BASC) involved in DNA damage detection and repair. This particular subunit dimerises with PMS2 to provide endonuclease capacity within the complex. MLH1 germline mutations give rise to HNPCC (hereditary non-polyposis colorectal cancer). Somatic mutations in this gene are important in sporadic colorectal cancers. Mutations of MLH1 lead to a mutator phenotype often manifested by microsatellite instability.

MSH2

MSH2 is a tumour suppressor gene, also involved in mismatch repair. It resides within the 'BRCA1-associated genome surveillance complex' (BASC) which detects and repairs DNA damage. MSH2, in complex with MSH6, forms a sliding clamp which traverses the DNA backbone detecting mismatched bases. MSH2 germline mutations also give rise to HNPCC. Similar to MLH1, somatic mutations in MSH2 are found predominantly in colorectal cancers. Mutations of MSH2 lead to a mutator phenotype often manifested by microsatellite instability.

CDC73

CDC73 (HRPT2) is a tumour suppressor forming part of the PAF protein complex, which is associated with RNA polymerase II and may therefore be involved in both initiation of RNA synthesis and RNA elongation. Mutations in this gene have been identified in tumours of the parathyroid, most often causing the endocrine disorder hyperparathyroidism (with or without jaw tumour).

MAP2K4

MAP2K4 is one part of the mitogen-activated protein kinase (MAPK) pathway, a signal transduction cascade which mediates certain extracellular signals via RAS/RAF resulting in transcriptional control of a wide range of genes. The MAP2K family of peptides regulate MAPK activity by phosphorylation. MAP2K4 mutations appear involved in many tumour types.

Gene	Samples	Experiments	Mutations	Unique Mutations	Papers
MLH1	1328	1325	44	38	25
MSH2	1306	1304	36	33	23
CDC73	278	272	39	32	11
MAP2K4	1557	1559	22	19	9

EWSR1/FLI1 Gene fusions

Ewing's sarcoma is a rare bone tumour, infrequently of extraskeletal origin, most frequently occurring in teenage children. The majority of these tumours contain a t(11;22)(q24;q12) translocation which fuses the EWSR1 gene on chromosome 22 with the FLI1 gene on chromosome 11. We have now curated the existing literature describing fusions between this gene pair.

Genes	Mutant Samples	Papers	Unique Mutations
EWSR1/FLI1	1133	115	28

The following curated genes have been updated for this release: CDKN2A, PTPN11, NRAS, MLH1, MSH2, KRAS, JAK2, MAP2K4, BRAF, HRAS, CTNNB1, MEN1, NF1, APC, VHL, EGFR, FLT3, PTCH, MPL, WT1, RET, CDC73, RUNX1, EWSR1, FLI1.

Web site upgrade

Genomic co-ordinates for individual mutations are now available in the data export section, together with the datasheets in the FTP site.

CGP Trace Archive

The CGP trace archive has been updated to contain all the sequencing traces used in our analysis of the samples and genes presented in the CGP Resequencing project (COSMIC red pages). The number of traces available for download is now approaching 9.5 million. The Archive itself has also been upgraded, so that it receives daily updates of CGP sequencing traces as they pass through our sequencing pipeline. Daily updates are available as separate files; these will be integrated into the main download files once per week.

Samples with trace data	Total number of traces available
276	9465645

COSMIC Statistics

Experiments	991743
Tumours	250869
Samples	251847
Mutant Samples	50949
Mutations	53098
Unique Mutations	10779
Papers curated	5449
Genes	4763
Fusions	1957

COSMIC 34

2007-11-08T16:59:59Z

This release of COSMIC includes the addition of BRCA1, BRCA2, and EWSR1/ERG gene fusion from the scientific literature. The website has been enhanced with an update of old gene names and the addition of further links (NCBI Entrez Gene, CCDS, Swiss-Prot and TrEMBL). The CGP Trace and Genotype Archive holding the groups sequence traces and genotype data is also now available.

Literature Curation

BRCA1 and BRCA2

BRCA1 and BRCA2 are tumour suppressor genes initially identified as inherited cancer susceptibility genes for breast and ovarian cancer. Both proteins been shown to have roles in genome surveillance, detection of DNA damage and its subsequent repair. However, they associate with different DNA repair complexes and generate different tumour histologies and spectra. Somatic mutations of either gene are rare, with BRCA2 being more frequently found to have somatic mutations, particularly in ovarian and pancreatic carcinomas.

We report that mutations in these two genes have been discovered at fairly low frequencies (2-3%), with BRCA2 mutated in a wider tissue range than BRCA1.

Gene	Unique Samples	Samples	Experiments	Mutant Samples	Papers	Mutations	Unique Mutations
BRCA1	1106	1106	1114	25	22	25	23
BRCA2	1142	1146	1145	29	16	33	29

EWSR1/ERG fusion

Fusions of EWSR1 and ERG are common events in skeletal (and the rarer extraskeletal) Ewing's Sarcoma. These fusions, found at a frequency of approximately 10% in bone tumours result from complex rearrangements, since the two partner genes are not transcribed in the same chromosomal direction.

Genes	Mutant Samples	Papers	Unique Mutations
EWSR1/ERG	77	49	11

COSMIC Data Updates

The CGP Resequencing screens and the following curated genes have received updates: BRAF, HRAS, CSF1R, CTNNB1, KIT, PDGFRA, PTEN, ACVR1B, ATM, ERBB2, BRCA1, BRCA2, KRAS, PIK3CA, FGFR2, ABL1, FGFR1, JAK2, SRC, STK11, CDKN2A, PTPN11, NRAS, FAM123B, APC, SMAD4, VHL, MSH6, MET, EGFR, FLT3, FBXW7, MEN1, NF1, RUNX1, FGFR3, RET.

CGP Trace and Genotype Archive

The groups sequence traces and genotype data are now available from the CGP Trace and Genotype Archive site. In order to access the data a Data Transfer Agreement must be completed and approved. A unique username and password will then be provided to access this resource.

Samples with trace data	276
Samples with genotyping data	1,135
Total number of traces	7,254,445

Gene Name Update

244 genes had their names updated (5.2%). It is still possible to search by the old gene name.

Website Upgrades

There has been an addition of several external gene links on the gene summary page. This includes links to NCBI Entrez gene, CCDS, Swiss-Prot and Trembl.

The sample summary page now also contains sample source information.

General Statistics

Experiments	984673
Tumours	246369
Mutant Samples	50032
Mutations	52146
Unique Mutations	10533
Papers curated	5271
Genes	4762
Fusions	685

COSMIC 33: Improved CGP data release

2007-09-05T11:59:52Z

The WTSI Cancer Genome Project (CGP) announces an updated data release policy. We will now be releasing confirmed somatic mutations on a bi-monthly basis. Confirmed and annotated somatic mutations identified in the previous two months will be released in COSMIC, continuing on at two-monthly intervals. Data will still appear within current COSMIC architecture of gene family/gene set and under appropriate studies. This new policy will result in expedited pre-publication release of curated somatic mutations as they are identified.

This new data will be available in the COSMIC blue pages, but will be most noticeable in COSMIC's CGP resequencing studies site (red pages), as this distinguishes CGP data from the literature curation.

CGP resequencing data is broadly divided (in the red pages) into 3 categories, 'Kinase', 'Pilot' and a new project, 'Renal'. Whilst the Kinase data is completed and published, the other two studies are much larger and still in progress. A collection of approximately 4000 genes has been selected for resequencing in a set of 40 matched pair cell lines ('Pilot' project) and 96 primary clear cell renal cancers. Each tumour sample in these projects has a matched normal sample, which allows the distinction of somatic mutations from germline variants. The pilot project currently comprises 1865 somatic sequence changes, whilst the Renal project, although less advanced than the Pilot, has identified 84 mutations to date. These will be automatically updated with all our confirmed data every bimonthly release.

Literature curation

RUNX1 (AML1) has been fully curated

RUNX1 is one subunit of the PEBP2 transcription factor, binding to DNA at enhancer sequences. This gene is one of the most frequent targets of chromosome translocations associated with leukemia. Small somatic mutations have also been observed, most frequently in myeloblastic leukaemia types (Acute myeloblastic Leukaemia, MyeloDysplastic Syndrome) and it is these that we have curated in COSMIC. Our data suggests a somatic mutation rate of approximately 10% in this phenotype.

Curated Gene Update

The following curated genes have received updates from the literature: APC, ATM, BRAF, CDH1, CDKN2A, CTNNA1, CTNNB1, CYLD, EGFR, ERBB2, ERG, ETV1, FBXW7, FGFR3, FLT3, GATA1, HRAS, JAK2, KIT, KRAS, MADH4, MPL, MSH6, NF1, NF2, NOTCH1, NPM1, NRAS, PIK3CA, PTCH, PTEN, PTPN11, RB1, RET, SMARCB1, SMO, SOCS1, STK11, SUFU, TMPRSS2, VHL, WT1, WTX.

General Statistics

This release includes 1563 new mutations identified in the set of 4799 genes; 1495 genes are new this month.

Experiments	968416
Tumours	239766
Mutant Samples	48959
Mutations	51054
Unique Mutations	10390
Papers curated	5103
Genes	4799
Fusions	445

COSMIC v32

2007-08-08T16:59:01Z

This release includes four new tumour suppressor genes and improved availability in Ensembl.

New external integration: Ensembl

We are continually striving to improve the utility of the data in COSMIC by integrating it closely with external resources. In this release, we provide a much closer integration with the Ensembl genome browser than previously. All our gene & mutation data now have location coordinates on the NCBI36 genome sequence, allowing us to use Ensembl "DAS" technology to display this information within their genome browser, aligned with their standard genome annotations. We have made this easily available, via a single link from our pages.

Literature curation

Four new tumour suppressor genes have been introduced to COSMIC this month, all receiving full literature curation of their somatic mutation data.

NF1

Neurofibromatosis is a familial disease with a complex phenotype including tumours of the central nervous system, caused by mutations in the NF1 tumour suppressor gene. Somatic mutations in tumours have also been identified in this gene, and it is these that we have fully curated.

NF2

The central form of neurofibromatosis is a similar familial central nervous system tumour syndrome, caused by mutations in the NF2 tumour suppressor gene. Somatic mutations in tumours have also been identified in this gene, and it is these that we have fully curated.

SOCS1

SOCS1 downregulates cellular cytokine signalling by its direct interaction with JAK1. It was first implicated in cancer after aberrant methylation was observed to inactivate its activity causing Hepatocellular Carcinoma. Somatic mutations have also been observed which inactivate this tumour suppressor and these have been curated.

TCF1

TCF1 binds to the promoters of several (largely liver-specific) genes, to enhance their expression. Somatic and germline mutations in this gene have been found which cause liver adenomas, and we have curated the somatic component.

The following curated genes have received updates from the scientific literature: KRAS, PIK3CA, JAK2, BRAF, HRAS, KIT, PDGFRA, PTEN, CDKN2A, VHL, EGFR, FBXW7, MEN1, RET

General Statistics for this release

Experiments	521624
Tumours	235207
Mutant Samples	47470
Mutations	49491
Unique Mutations	9699
Papers curated	5053
Genes	3304
Fusions	445

COSMIC (v31) now includes Gene Fusion Data

2007-07-27T14:59:27Z

The CGP COSMIC team is pleased to announce the addition of gene fusion/translocation somatic mutation data from the literature to the database. Currently, the census of known cancer genes is dominated by somatically generated fusion genes that have been identified primarily in leukaemias, lymphomas and soft tissue tumours. Until now, we have concentrated on curating somatically point mutated cancer genes for COSMIC. Almost all known cancer genes that have somatic point mutations are, however, now curated in COSMIC. In the coming months we will therefore be searching the scientific literature and annotating genes involved in gene fusions and their partners for addition into the COSMIC database.

We have launched this new facility, complete with new views for this data type, with the curation of TMPRSS2, a gene frequently found to be fused to ETS family transcription factors in adenocarcinoma of the prostate. These mutations have served to spur increased investigation into the potential role of fusion genes in adult solid tumours. The move to curate fusion genes is an important addition and will further enhance COSMIC as the most comprehensive source for somatic mutation data from human cancers.

Fusion Gene Pairs

TMPRSS2/ETV1

TMPRSS2/ERG

Website Upgrades

The fusion data has been integrated into existing pages and overviewed in new pages: Translocations Overview and Translocations Summary.

This new data can be viewed graphically and textually.

The image above shows the table of inferred breakpoints (determined from a sample's observed fusion mRNA spectrum) for a fusion gene pair.

The image above shows a graphical representation of the observed mRNA transcripts from which the inferred breakpoints are calculated.

Further information of the new gene fusion website features is available in the help pages.

Genes from Literature Curation

A new homepage has been created for genes which have received full curation of the scientific literature. This is a new page which allows the distinction of these genes from CGP's data release, for which no literature has been curated.

Curated Gene Update

The following curated genes have also received updates from the scientific literature: CDKN2A, GATA1, NOTCH1, NPM1, NRAS, JAK2, KRAS, PIK3CA, BRAF, HRAS, CEBPA, CSF1R, CTNNB1, KIT, PDGFRA, PTEN, RB1, MET, EGFR, FLT3, WT1, APC, MADH4, FBXW7, FGFR3.

General Statistics for this release

Experiments	515535
Tumours	230057
Mutant Samples	46978
Mutations	48911
Unique Mutations	9014
Papers curated	4938
Genes	3302
Fusions	438

COSMIC v30

2007-06-06T14:01:00Z

Today we release full literature curations of five tumour suppressor genes MEN1, ATM, CYLD, FBXW7, WTX; 4712 samples were examined in 112 papers, recording 468 mutations. Additionally, we release two new CGP resequencing studies which add a further 91 new genes to COSMIC.

Literature Curation

Curation of the scientific literature has been completed for five new genes from the cancer census. All five genes are tumour suppressors, causing phenotypes via their inactivation:

MEN1 (Multiple Endocrine Neoplasia Type 1)

Somatic mutations in this gene have been found in tumours from several endocrine sites, recapitulating those seen in patients carrying germline mutations including tumours in the pituitary, pancreas and parathyroid. MEN1 encodes a nuclear protein thought to be a transcriptional regulator.

CYLD (Cylindromatosis)

This gene has been found to have mutations in sporadic cylindromas, tumours arising from skin adnexal structures (such as hair follicles and glands), principally on the face and scalp. CYLD encodes a deubiquitinating enzyme regulating cell signalling including the NF-kappaB pathway.

FBXW7 (CDC4)

Mutations inactivating FBXW7 have been found in a range of cancer types including colorectal, ovarian and T-ALL. The protein is involved targeting a number of key proteins, including NOTCH1 and MYC, for ubiquitin-mediated degradation.

ATM

This gene encodes a protein kinase involved in cell cycle checkpoint control. Amongst other key cell cycle components, it has been shown to phosphorylate TP53 and CHEK2 in response to DNA damage. Germline mutations causes Ataxia-telangiectasia (AT) a recessive disorder characterized by cerebellar ataxia, telangiectases, immune defects, and a predisposition to malignancy, primarily lymphoid in origin.

WTX (FAM123B)

Recently discovered, WTX is inactivated in approximately 30% of Wilms Tumours. Located on the X chromosome, this tumour suppressor only requires a 'single-hit' for tumourigenic inactivation.

New tumour suppressor gene statistsics:

Gene	Samples	Experiments	Mutations	Papers
MEN1	1680	1683	196	66
ATM	1714	1692	198	33
FBXW7	1207	1204	60	10
WTX	82	82	7	1
CYLD	29	29	7	2

The following curated genes have also received updates: BRAF,HRAS,CEBPA,CTNNB1,KIT,PDGFRA,PTEN,SMARCB1,ERBB2,JAK2,CDKN2A,PTPN11,NRAS,KRAS,PIK3CA,APC,CDH1,MADH4,EGFR,FLT3,MPL,WT1,FGFR3

CGP resequencing studies

91 new genes have been examined in our pilot set of matched pair cell lines, resulting in the discovery of 22 new mutations:

Study	Genes	Experiments	Samples	Mutations
Integrin alpha family	16	640	40	11
Miscellaneous genes of interest from literature sources	75	3000	40	11

General Statistics for this release

Experiments	499958
Tumours	217944
Mutant Samples	44491
Mutations	46364
Unique Mutations	8855
Papers curated	4794
Genes	3302

COSMIC v29 released

2007-05-09T09:07:48Z

COSMIC release 29 includes 22 new CGP resequencing studies, comprising 567 new genes within which 192 new mutations have been identified. Additional updates to our curation of the scientific literature have also been included, adding a total of 1041 mutations to this release.

CGP Resequencing Studies

567 genes have been examined in our pilot set of matched pair cell lines:

2

Study	Genes	Mutations
PAX transcription factor family	11	5
Tripartite motif-containing protein family	56	28
Genes on APC/CTNNB1 pathway	59	25
FK506/rapamycin binding protein family	26	5
Diacylglycerol kinases and other lipid kinases	18	17
SMAD protein family	24	7
Histone acetyltransferase	7	2
Dual specificity phosphatases	23	2
Genes associated with ERB family of RTKs	8	3
Genes associated with MYC proteins	21	12
Ubiquitin specific peptidase family	50	16
C-X-C/C-C motif chemokine receptor genes	19	8
Essential For Cell Division - derived from a siRNA screen in human cells	21	5
Genes from RNAi TSG gene screen	5
Glycolysis associated genes	23	4
Integrin beta family	8	8
Small ubiquitin-like modifier (SUMO) protein family	14	2
14_3_3 family of scaffold protein	8	1
STAT and SOCS gene families	43	7
Serpin/TIMP peptidase inhibitor families	46	17
Sorting NeXin family	27	3
Genes associated with RAS proteins	53	13
Total new CGP data	567	192

Literature curation

89 new publications have been curated, updating the information for the following genes: JAK2,HRAS,CEBPA,PTEN,RB1,RET,ERBB2,CDKN2A,GATA1,NRAS,KRAS,PIK3CA,EGFR,CTNNA1,APC,CDH1,MADH4

General Statistics for this release

Experiments	482902
Tumours	206972
Mutant Samples	42266
Mutations	44062
Unique Mutations	8420
Papers curated	4515
Genes	3220

COSMIC v28 Released

2007-04-04T12:33:10Z

This months COSMIC release comprises a substantial increase in the CGP resequencing data, adding 1033 new genes to the system, together with updates to the scientific literature curation.

CGP Resequencing Studies

26 new studies have been included in this release, containing 1033 new genes which have been examined through the pilot matched pair cell line set.

Study	Genes	Mutations
ADAM metallopeptidase family	40	27
Cyclins and Genes associated with RB	68	20
Nfkappa signalling family	58	14
Phospholipase C Family	13	16
Protein Kinase anchor proteins	32	15
Ral Guanine nucleotide dissociation factors	6	3
Hypoxia inducible factor pathway	23	11
SerThr Phosphotases (PPP)	69	17
Integrin Binding proteins	27	1
K homology RNA-binding domain, type I	25	9
Cytochrome C oxidase family	24	3
DNA methylation and histone deacetylation	38	10
Heat shock proteins	81	20
Ets transcription factor family	28	5
High Mobility Group proteins	24	2
Immediate early/regulator of G-protein signalling family	25	5
Kallikrein protease family	16	5
Matrix metallopeptidase family	21	7
Genes implicated in stem cell regulation	63	12
TCA cycle genes	56	16
Forkhead transcription factor family	43	11
TP53 responsive genes	76	22
Ubiquitination pathway genes	63	21
Ubiquitin Ligases	72	36
DEAD Box proteins	60	25
Genes associated with TP53 and targets	47	16

Curated Gene Update

The following fully curated genes also received minor updates : APC, BRAF, CDKN2A, CTNNB1, EGFR, ERBB2, HRAS, KIT, KRAS, NOTCH1, NPM1, NRAS, PDGFRA, PTEN, PTPN11, RB1, WT1.

General Statistics for this Release

Experiments	455765
Tumours	204457
Mutant Samples	41259
Mutations	43021
Unique Mutations	8122
Papers curated	4426
Genes	2671

COSMIC v27 released

2007-03-14T10:22:52Z

This months release of COSMIC comprises upgrades to both the web site (which now allows searching by gene/sample name or keyword) and data, with new CGP resequencing studies and curated genes. COSMIC now contains data on over 200,000 tumour samples and 400,000 individual experiments. Of these 202109 tumours, 40331 were found to contain one or more mutations (19.9%).

CGP Resequencing Studies

Two new studies examine our pilot data set comprising 40 cancer cell lines that have a matched normal cell line, allowing all of the mutations to be confirmed as somatic.

Notch signalling proteins

This group of proteins comprise the Notch receptors and other proteins which are involved in Notch signalling. The Notch signalling pathway allows cells to communicate with each other and plays a crucial role in developmental regulation. NOTCH1 mutations have been associated with T-cell acute lymphoblastic leukaemia.

Phosphatidylinositol metabolism

This gene set includes proteins which control the synthesis and turnover of phosphatidylinositol which is synthesised in the endoplasmic reticulum before translocating to cytosolic membrane surfaces where it plays an important role in many cellular processes including cell signalling. Mutations in the phosphatidylinositol-3-kinase PIK3CA and the lipid phosphatase PTEN are associated with many types of cancer.

Literature Curation

STK11 (LKB1)

STK11 is a tumour suppressor, physically associating with p53 to effect growth suppression via p53-dependent apoptosis pathways; restoring gene activity into cancer cell lines defective for its expression results in a G1 cell cycle arrest. It has been identified as the cause of Peutz-Jeghers syndrome, an autosomal dominant disorder inducing an increased risk of melanocytic macules, gastrointestinal polyps and various neoplasms.

STK11 Statistics

Samples	2344
Mutations	92
Unique Sequence Changes	63

WT1

Wilms tumour is a solid cancer usually occurring in childhood, caused by malignant transformation of renal stem cells retaining embryonic differentiation potential. Several tumour suppressor genes have been associated with the development of WT, most classically the WT1 zinc finger DNA binding protein located at chromosome 11p13. A number of isoforms of the transcription factor WT1 exist, unusually exerting control over expression of target genes during both their transcription and splicing.

WT1 Statistics

Samples	1710
Mutations	106
Unique Sequence Changes	68

Website Upgrades

Search Facility

A major update to the COSMIC website this month is the Exalead search facility, allowing for easier navigation of the site. In the 'Text Search' field on the home page, you can search for a gene name or accession number, a sample name or id, or a tumour primary site or sub-site. There is a help page for more advanced searches, which can be accessed by clicking on the question mark in the search box, or the help button in the sidebar.

General Statistics for this release

Experiments	408164
Tumours	202109
Mutant Samples	40331
Mutations	42057
Unique Mutations	7736
Papers curated	4348
Genes	1638

COSMIC v19 released

2006-06-07T09:34:34Z

This month's release of COSMIC includes the Cancer Genome Project screen of the GAP-GEF gene set and new information displays.

GAP-GEF Screen

This gene set, consisting of 173 genes, is comprised of proteins that function to regulate the activity of proteins with GTPase activities. GTPase activating proteins (GAPs) promote hydrolysis of GTP-GDP. Guanine nucleotide exchange factors (GEFs) promote GDP/GTP exchange. Both classes modulate the function of the small monomeric GTPases (including the RAS oncogene family) and other key signalling proteins that use the conversion of GTP-GDP as a molecular switch to regulate function. This system of GTPase/GAP/GEFs regulates a wide variety of cellular processes including growth, differentiation, survival and motility.

Web Improvements

Zygosity and somatic/germline status information are now available for mutations in COSMIC, CGP Resequencing and Cancer Cell Project websites. The somatic/germline status is listed on the sample detail page and the export function with the following statuses:-

Not specified
Confirmed somatic mutant
Reported elsewhere as a somatic variant
Confirmed germline variant
Reported elsewhere as a germline variant
Variant of unknown origin

Zygosity information is available on the mutation detail page with the following statuses:-

Unknown
Homozygous
Heterozygous

General Statistics for this release

Experiments	264296
Tumours	155902
Mutant samples	27732
Mutations	28859
Papers curated	3419
Genes	1337

COSMIC v18 released

2006-05-04T09:15:38Z

The CGP Resequencing Studies Website is released this month, which will act as a repository for data from CGP resequencing efforts to identify novel somatic mutations in human cancer. The pages have their own distinctive red colour scheme to denote this. Prior data on sets of genes/samples systematically screened for mutations were previously integrated into the "blue" COSMIC pages. This will continue with data now being submitted, prepublication, to and held on the new site. This will allow users to browse, search and evaluate these data more effectively. The web resources that are now available are detailed below:-

COSMIC (Blue): All data screened from literature and CGP based projects.
CGP Resequencing Studies (Red): Somatic mutations from systematic large scale resequencing of genes in human cancers.
CGP Cancer Cell Line Project (Green): Resequencing of known cancer genes and other analyses of human cancer cell lines.

Curation of PTCH

37 papers from the scientific literature have been curated for the PTCH gene in this release. Adding an additional 897 experiments and 168 mutations.

General Statistics for this release (COSMIC)

Experiments	254544
Tumours	153528
Mutant samples	27426
Mutations	28534
Papers curated	3393
Genes	1176

COSMIC DAS track

Ensembl has recently moved to the NCBI 36 assembly of the human genome whilst COSMIC genes and mutations are currently mapped to build 35. This has caused some disparity with the COSMIC DAS track. Therefore we suggest only using the cosmic DAS track on the most recent ensembl archive site(http://feb2006.archive.ensembl.org/index.html).Provided below is a link that will open the appropriate website with the DAS source attached:

http://feb2006.archive.ensembl.org/Homo_sapiens/contigview?conf_script=contigview;c=7:139949999.5:1;w=200000;h=7;add_das_source=(name=COSMIC+url=http://das.ensembl.org/das+dsn=cosmic_ncbi_35+type=ensembl_location+color=blue+strand=b+labelflag=n+stylesheet=y+group=n+depth=0+score=n+active=1)

COSMIC v16 released

2006-03-08T14:16:29Z

Released for March are data from a kinase domain screen of malignant gliomas. These data cover approximately 400kb of sequence in each of 9 tumours, including data from recurrent/resistant tumours.

We have recently completed a screen for somatic mutations of the kinase domain encoding exons of the entire protein kinase family in a series of human malignant gliomas. The results are presented in this release of COSMIC. No commonly mutated kinase domain was found in these studies. However, as is the case with our other work in this area, deep sequencing data from human tumours is informative about the processes that have contributed to oncogenesis in the patient. Two gliomas recurrent after temozolomide (alkylator) chemotherapy, but not a third recurrent after XRT alone, had the highest mutation prevalence of any tumours we have analysed to date. These data suggests a link between mutation prevalence and recurrent/resistant brain tumours treated with alkylator chemotherapy.

Statistics

Experiments	235213
Tumours	143427
Mutant samples	25360
Mutations	26388
Papers curated	3207
Genes	1035

A COSMIC Expansion

2006-02-07T10:35:32Z

The Catalogue Of Somatic Mutations In Cancer is two years old and has mutation data for over 1,000 genes, curated from over 3,000 published papers and unpublished data from the Cancer Genome Project.

The original aim of COSMIC continues with the curation of somatic mutation information from the literature for known cancer genes. During 2005 data for 9 genes was collected; ABL1, CDKN2A, EGFR, GATA1, JAK2, MSH6, NOTCH1, PTPN11 and SMO. Genes that were curated in 2004 were updated as new data was published.

The number of genes in COSMIC expanded rapidly when the Cancer Genome Project at the Wellcome Trust Sanger Institute published 3 studies of somatic mutations in the protein kinase gene family (518 genes in total). This data provides a unique insight to the somatic mutations in breast, lung and testicular cancers.

More recently the Cancer Genome Project has been submitting unpublished somatic mutation data to COSMIC (link). The data comes from genes involved in apoptosis, DNA repair, maintenance and metabolism and the Inositol Polyphosphate Phosphatase and Heterotrimeric G-Protein families.

In another new departure the COSMIC software was used to create a new web site the Cancer Cell Line Project. This separate site, with it's own 'mint' colour scheme, contains the results from the sequence analysis of 14 known cancer genes in over 700 cancer cell lines. Initial sequence data for 4 genes analysed in the NCI-60 is also available. This work is in progress and more results will be posted in the coming months. What is more, the number of genes in this project will continue to increase; providing genetic data for this wide set of cancer cell lines.

There have been many enhancements to the web site over the past 12 months. A tissue overview provides a summary of mutations reported in a selected tissue. New pages were created to show more details of mutations and samples and give greater depth to the data. There are also links to other data such as genome copy number information.

COSMIC has been summarised in The British Journal of Cancer (Forbes et al, 2006).

This month sees the update of; BRAF, CDKN2A, EGFR, ERBB2, HRAS, KRAS, NRAS, PTEN, PTPN11 and SMARCB1. In addition the Cancer Genome Project has submitted unpublished data for genes involved in apoptosis.

There are plans to continue the development of COSMIC in terms of data content and data presentation. We are always happy to receive feedback and suggestions (email: cosmic@sanger.ac.uk).

Statistics

Experiments	228,669
Tumours	142,569
Mutant samples	25,176
Mutations	26,194
Papers curated	3,013
Genes	1,035

COSMIC v14 released

2006-01-10T15:42:40Z

The COSMIC team is proud to announce the release of COSMIC-14 with data for CDKN2A(p16) and more unpublished data from the CGP.

DNA REPAIR, MAINTENANCE AND METABOLISM

The Cancer Genome Project has released further unpublished somatic mutation data from a screen of 41 cancer cell lines. The 302 genes in this release are involved or associated with DNA repair, maintenance and metabolism. The genes can be viewed together or in 5 subgroups; Telomerase Complex, SWI/SNF, DNA replication, Nucleotide Metabolism and DNA Damage Response and Repair. In total 119 somatic mutations were identified in this study.

CURATION OF CDKN2A

CDKN2A (also known as p16) is a tumour suppressor. It induces cell cycle arrest by inhibiting the phosphorylation of Rb by the cyclin-dependent kinases CDK4 and CDK6. So far 453 papers have been curated for this gene with 2,591 mutations recorded from 16,883 samples.

STATISTICS

Experiments	219,037
Tumours	140,212
Mutant samples	24,817
Mutations	2,637
Papers curated	3,379
Genes	870

COSMIC v13 released

2005-12-13T10:54:42Z

Somatic mutation data from new gene families

In a major new departure the Cancer Genome Project is proud to release further somatic mutation data. The results from the sequencing of two gene families, Inositol Polyphosphate Phosphatases and Heterotrimeric G-Proteins, have been added to the data for the Protein Kinase genes . This data will be expanded in the future with the addition of further gene sets.

Updates to existing genes

Nine genes in COSMIC have been updated with further data; NRAS, RB1, ERBB2, HRAS, PTEN, TP53, KRAS, APC and CDKN2A

New DAS data source

The Cancer Genome Project is pleased to announce the release of a DAS source devoted to the genes and mutations within COSMIC. Using this source you will be able to view the genes and mutations from COSMIC within a genome browser or the DAS client of your choice.

All 587 genes in COSMIC are exported as features. Each of these features displays the genomic 'footprint', which encompasses both exonic and intronic sequence between the start and end points of the CDS sequence. A link is attached to each feature, providing a mechanism for the client to link back directly to the gene entry on the COSMIC website.

In addition to the gene footprints, there are also a large number of unique mutations. These are also displayed as features, with links back to the mutation summary page in COSMIC. The database currently holds 2812 unique mutations, of which 1035 are currently exported. This subset is comprised of all the single nucleotide substitutions. More complex mutations will be included, as the genomic coordinates are mapped.

The DAS source can be found at the following URI:

http://das.ensembl.org/das/cosmic_genomic

The easiest way to view this source is to place the following URI in your browser:

http://www.sanger.ac.uk/turl/6d8

This will attach the DAS source and display some of the mutations found in BRAF. Additional configuration can be performed on the track, by clicking on the track name. For more information, see the help pages on the Ensembl website.

COSMIC statistics

Experiments	190,576
Tumours	124,381
Mutant samples	23,232
Mutations	2,228
Papers curated	2,812
Genes	587

Protein kinase mutations in lung cancer

2005-09-01T15:03:19Z

The Cancer Genome Project has sequenced all protein kinase genes in lung cancer - the most common cause of cancer deaths worldwide

There are over 27,000 new cases of lung cancer in the United Kingdom each year. Protein kinases are frequently mutated in human cancer and inhibitors of mutant protein kinases have proven to be effective anticancer drugs. The Cancer Genome Project has screened the complete coding sequence of all 518 protein kinase genes in 33 lung cancers. This study, published in Cancer Research, is the largest survey reported to date of somatic mutations in lung cancer.

The Cancer Genome Project at the Wellcome Trust Sanger Institute was established in 2000. Its goal is to identify mutations that occur in cancer cells to enable the development of new diagnostics and new treatments and advance our understanding of the biology of cancer.

The Wellcome Trust Sanger Institute Cancer Genome Project and their collaborators have published the latest results of their survey of genes that might be implicated in cancer. The report is published in Cancer Research on Thursday 1st September 2005 and is also available through COSMIC.

The gene set chosen was a class called protein kinases, key controllers of cell growth and death. Members of this family have been shown to be important in cancer. However, the whole set has never been sequenced in a single set of lung tumours. The study generated over 40 million bases of DNA sequence (1.3 million for each sample).

This work identified 188 somatic mutations in 141 protein kinase genes. There was considerable variation in the number of mutations found in each tumour. The results indicate that several mutated protein kinases may be contributing to lung cancer development, but that mutations in each one are infrequent. Larger studies are warranted to further explore these initial findings. Cancer is a complex set of diseases that will affect 1 in 3 people. This work in the CGP is but one part of a global effort to further understanding of cancer and move towards better diagnosis and treatment.

COSMIC Website Update

2005-08-03T15:11:33Z

The COSMIC web site has been updated with additional data from the literature and unpublished data from the Cancer Genome Project.

SOMATIC MUTATION DATA FOR KNOWN CANCER GENES

Data for 3 genes has been curated from the literature and included in COSMIC; ABL1, GATA1 and SMO.

SOMATIC MUTATIONS OF THE PROTEIN KINASE GENE FAMILY

The screen of the protein kinase gene family by the Cancer Genome Project now includes two new tumour types; lung cancer and testicular germ-cell tumours. There are marked differences in the mutation prevalence between these two tumour types.

CANCER CELL LINE PROJECT

The mutation data for 9 further genes has been included on the web site giving a total of 550 mutations. The genes are APC, CDH1, CTNNB1, HRAS, MADH4, PIK3CA, PTEN, RB1 and STK11. The sequencing of these genes is not necessarily complete but the cell lines with mutations have been confirmed and the experiments will continue to finish this work.

COSMIC STATISTICS

179,563 Experiments

118,134 Tumours

22,005 Mutations

2,090 References

534 Genes

The Cancer Cell Line Project

2005-06-14T10:51:01Z

The Cancer Genome Project has released data from a systematic characterisation of the genetics and genomics of over 600 cancer cell lines including the NCI-60.

For decades, human immortal cancer cell lines have constituted an accessible, easily usable set of biological models with which to investigate cancer biology and to explore the potential efficacy of anticancer drugs. Nevertheless, cancer cell lines have been criticised because they may represent a highly selected subgroup of the cancer classes from which they have been derived and have acquired additional genetic abnormalities in vitro.

Prior knowledge of the genetic abnormalities in cancer cell lines may, however, allow more informed choice of cell lines in biological experiments and drug testing and better interpretation of results. The aim of these studies, therefore, is to systematically provide this information and hence to improve the utility of cancer cell lines for the cancer research community.

We have assembled a large series of cancer cell lines from major publicly accessible repositories around the world. From this series we have currently selected a series of more than 600 lines for detailed analysis. This series has been designed to encompass a broad range of tumour types. It includes most cell lines that have been used extensively in cancer research, including the NCI-60 set.

The analyses that have been performed so far are;

Sequencing of known cancer genes

The coding exons and immediate flanking intron sequences of genes from the Cancer Gene Census are being sequenced. These studies are ongoing and the current release includes BRAF (34 mutations), KRAS (79 mutations) NRAS (46 mutations) and TP53 (216 mutations).

Copy Number analysis

We have analysed the cancer cell lines for genome-wide loss of heterozygosity and copy number changes using Affymetrix 10k SNP DNA microarrays.

Microsatellite instability

The cancer cell lines have been assessed for microsatellite instability using the BAT markers (BAT25, BAT26, D5S346, D2S123 and D17S250).

Cell line ancestry

The genotypes generated from the Affymetrix 10k SNP DNA microarrays have been used to compare the genome-wide genotypes of all the lines with each other. This has identified lines with very similar genotypes suggesting that they originate from a common ancestor and are probably derivatives of each other.

Protein kinases and breast cancer

2005-05-23T09:52:34Z

The Cancer Genome Project has screened the complete coding sequence of all 518 protein kinase genes in 25 breast cancers. This study, published online in Nature Genetics, is the largest survey reported to date of somatic mutations in cancer.

The Wellcome Trust Sanger Institute Cancer Genome Project and their collaborators have published the latest results of their survey of genes that might be implicated in cancer. The report, published in Nature Genetics online on Sunday 22 May 2005, identifies a new \x91mutator signature\x92 and shows that there is no common mutation of a set of key genes in the 25 breast cancer samples examined.

The gene set chosen was a class called protein kinases, key controllers of cell growth and death. Members of this family have been shown to be important in cancer. However, the whole set has never been sequenced in a single set of tumours. The data, 1.3 million bases of DNA sequence for each sample, is the most that has ever been obtained from a single tumour type.

The study identified a novel mutation pattern that had never been seen before and gives new insight to the number and diversity of mutations in cancer. There was no indication that any of the protein kinase genes was frequently mutated in this set of breast cancer samples. Cancer is a complex set of diseases and this one large study is only part of a global effort to unlock the genetic secrets of a disease that affects 270,000 people each year in the UK alone.

COSMIC's first anniversary

2005-02-04T13:17:26Z

The COSMIC database and web site have been updated and now have somatic mutation data from 21 genes.

New Data

CEBPA is mutated in 7% of haematopoietic and lymphoid tissue tumours. It arrests cell proliferation by inhibiting the kinases CDK2 and CDK4.
CTNNB1 or beta-catenin is mutated in a variety of tumours. The gene encodes an adherens junction protein that is critical for the establishment and maintenance of epithelial layers
KIT is characterised by two clusters of mutations in and around the kinase domain of the gene with frequent mutations in haematopoietic and lymphoid tissue tumours (19%) and soft tissue tumour (32%).
PTEN has mutations through the whole coding sequence with a hot spot at codon 130. Tumours of the central nervous system and endometrium frequently have mutations in this gene (19% and 34% respectively)
SRC is homologous to the v-src gene of the Rous sarcoma virus and has one mutation that has been found in 10 samples.
SUFU encodes a component of the sonic hedgehog/patched signaling pathway and is mutated in central nervous system tumours.

Statistics

21 genes

104,682 tumours

18,478 samples have mutations

1,755 unique mutations

1,672 papers have been curated

Census Update

2005-01-06T10:51:52Z

The Cancer Gene Census is continually reviewed and updated when inadvertent omissions are brought to our attention, gene status changes or new cancer genes appear in the literature. We would like to report the addition of 43 genes to the census; the majority of which are involved in rare translocations in leukaemias and lymphomas. As well, several uncommon translocation partners for solid malignancies have been added. We are grateful to Professor Felix Mitelman for pointing out these omissions and providing information for updating the census.

COSMIC Update

2004-12-17T09:09:50Z

The COSMIC team is proud to release somatic mutation data for CSF1R, RB1, RET and SMARCB1. This information has been curated from the scientific literature. Somatic mutation data from 15 genes can be queried and viewed through the COSMIC web site.

Data

The 4 new genes in COSMIC give data on specific tumour types and increase the breadth of information that can be queried and displayed.
CSF1R, also known as the oncogene FMS, is a receptor kinase that is mutated in ~5% of myelodysplastic syndrome cases. Mutations in this gene have been associated with a predisposition to myeloid malignancy.
RB1 is mutated in more than 11% of the tumours that have been studied. It is frequently somatically mutated in cases of retinoblastoma (47%) while germline mutations predispose to the same disease.
RET, a tyrosine kinase receptor, is somatically mutated in 38% of thyroid medullary carcinomas. Germline mutations in the RET gene are associated with multiple endocrine neoplasia, type IIA and type IIB, medullary thyroid carcinoma and Hirschsprung disease.
SMARCB1, also known as SNF5/INI1, is frequently somatically mutated in soft tissue rhabdoid tumours (41%). These are highly malignant cancers that usually occur in young children.

Statistics

15 genes

73,767 tumours

13,420 samples have mutations

536 unique mutations

1,104 papers have been curated

COSMIC Update

2004-11-12T16:59:48Z

The COSMIC team are proud to include somatic mutation data for FGFR2, FGFR3, FLT3, MET, PDGFRA and PIK3CA on the COSMIC web site.

Data

The number of genes with data in COSMIC has more than doubled in this release of the database. The additional data represents a set of genes that have a lower, but nevertheless important, mutation frequency in human cancer as a whole. In specific malignancies genes such as FLT3 do have a significant role as can be seen from the data collected in COSMIC.

Gene	Number of analysed samples	Number of samples with mutations
BRAF	5158	736
ERBB2	714	8
FGFR2	30	2
FGFR3	1735	481
FLT3	7610	1499
HRAS	11876	477
KRAS2	35716	8302
MET	1081	59
NRAS	13884	1132
PDGFRA	146	25
PIK3CA	396	89
TOTAL	78346	12810

Number of unique mutations 307

Number of curated papers 976

Website Changes

Home Page

We have added a link to an ATOM feed for those people with ATOM enabled news feed readers. Adding this link to your feeds list will allow you to see the latest news from the COSMIC site as and when it is available.

Distribution View

A totals column has been added to the 'Details' table to show the total number of mutated samples that are listed.
Links to show only negative data have been added to the'More Details' links in the 'Details' table.
The Insertions and deletions table has been split to show different information for the two types of mutation.

Gene Selection

Genes can now be selected by chromosome, from the karyotype graphic, or as always from an alphabetical list.

References Page

The complete list of references for a specific gene can now be exported in a variety of formats including Excel.

Mutation Data Page

All pages with samples containing more than 100 samples have been split to reduce their size. However, the export function will still export all the samples as selected.

COSMIC Update

2004-09-29T22:44:48Z

We are pleased to announce an update to the COSMIC website. To coincide with the nature paper on ERBB2 we have added all the data for this gene to COSMIC. There have also been a number of improvements to the interface that we hope you will find useful.

New Data

ERBB2

Today, Nature publish our recent findings, the first description of small intragenic ERBB2 mutations in human cancer. Primarily found in non-small cell lung adenocarcinomas, the mutations identified are suggestive of inappropriate activation of ERBB2 kinase activity.

This addition brings 8 new mutations and 714 new samples to the database. Increasing the total number of mutant samples to 10655 and the total number of samples to 58032.

Website Changes

Distribution View

The summary table has been removed in favour of a new gene summary page. Containing all the data from this table, plus much more.
The mutations tables have been expanded to show insertions, deletions and complex mutations.
Information about the negative samples is now available and can be viewed by clicking on the 'More Details' link in the Details table. Like the positive samples, this data can also be exported in various formats.
A new insertions and deletions track has been added to the main image. This will allow us to display a larger number of genes with more complex mutation sets.
A complex mutations track has also been added to display those mutations (multiple base substitutions) which don't quite fit into any of the other categories.

Gene Summary Page

This has grown from the original four row summary table, on the distribution page, into a full page overview of the information stored about a specific gene.

Mutation hot spots: The mutation summary shows those areas of the transcript that have a high density of mutations. This can be used to go directly to the area of interest on the mutation distribution view.
References: A quick glance will show the most recently published paper that was analysed by the COSMIC staff.

Sample Summary Page

Here you will find a page containing all the information about a particular sample. Some of the previously unavailable information, such as details about the individual, has been been made available.

Genes Tested: Quickly identify all the genes in COSMIC that have been tested against the selected sample.
References: Locate all the references that have included the sample.

Reference Summary Page

For the first time in COSMIC you can see all the samples from one paper in one location. In addition to this there are also details about the genes screened and the mutations that were found.

Loss of heterozygosity maps

2004-08-17T10:41:55Z

The Cancer Genome Project has released whole genome loss of heterozygosity maps from over 800 cancer cell lines and normal samples.

Tumour suppressor type cancer genes are frequently inactivated in tumours by a small mutation covering one or a few base pairs in one allele and a large deletion in the second allele often removing many megabases. Polymorphic markers that are heterozygous in the germline are rendered homozygous in the deleted region giving rise to a loss of heterozygosity (LOH). Examples of cancer genes that have been found in regions of LOH include RB1 on chromosome 13 and p53 on chromosome 17.

The Cancer Genome Project at the Wellcome Trust Sanger Institute has typed 395 polymorphic CA/GT repeats in a total of 832 samples, 677 cancer cell lines and 155 normal samples to identify regions of LOH. For most of these lines there is no matched normal DNA however it is possible to predict likely regions of LOH based on runs of consecutive homozygous genotypes. In our data 99 out of 155 lung cancer cell lines have 6 or more consecutive homozygous typings on chromosome 17 (in addition to other regions of LOH across the genome). This compares with 0 out of 127 unrelated normal samples that have 6 or more consecutive homozygous typings on chromosome 17. We have used the LOH data to identify regions of the genome, which might harbour tumour suppressor genes, for further analysis.

The LOH map interface lets you select a chromosome and tumours and vary the number of consecutive homozygous typings to reveal potential regions of LOH. The data from across the genome can also be displayed. We have compared the genotyping data from the cell lines to identify synonymous lines with similar genotypes that are likely to have been derived from the same person. Furthermore you can view the names of all of the cell lines we have analysed.

The LOH Maps have;

Genotype data from 677 cancer cell lines
Genotype data from 155 normal samples (28 cell lines, DNA from 127 peripheral blood samples)
A total of 307,052 successful genotypes.