Written and maintained by CASRAI Editorial Board
Last updated
Who searches this: cancer-biology researchers and students who meet CCLE in a paper or a gene-expression query and need to know what it is, what data it holds and how to cite it correctly.
What is the Cancer Cell Line Encyclopedia (CCLE)?
CCLE is the Cancer Cell Line Encyclopedia: a large, publicly described resource of genomic and pharmacologic data on human cancer cell lines. The Broad Institute describes it as a collaboration between the Broad Institute and the Novartis Institutes for Biomedical Research, begun in 2006, with the goal of characterising human cancer cell lines genetically and pharmacologically and linking drug sensitivity to genetic and lineage patterns.
The phrase is used in three ways, and papers rarely say which: the cell line panel (the lines themselves), the CCLE dataset or database (molecular and drug-response data generated on that panel), and the project (the Broad/Novartis collaboration and its publications). Searches for ccle database and ccle cell line are usually about the second and first of these.
Who runs CCLE and what it covers
The Broad CCLE site organises the work in three phases and ties each to specific publications. In the original 2012 Nature paper, 947 human cancer cell lines spanning 36 tumour types were genomically characterised, and about 480 of them (479 in the abstract) were profiled for response to 24 anticancer drugs, each tested in an 8-point dose-response format. The 2019 Nature paper extended characterisation to more than 1,000 lines.
CCLE timeline by phase and publication
| Phase | Publication (as listed on the Broad CCLE site) | What it added |
|---|---|---|
| I | Barretina et al., Nature 2012 (483:603-607) | 947 lines genomically characterised; pharmacologic profiling of about 480 lines against 24 drugs |
| I | Nature 2015 (pharmacogenomic agreement) | Listed by Broad as a phase I publication on agreement between pharmacogenomic datasets |
| II | Nature Genetics 2013; Ghandi et al., Nature 2019 (569:503-508); Nature Medicine 2019 | The 2019 paper expanded to more than 1,000 lines with additional data types; the Nature Medicine paper concerns metabolism |
| III | Cell 2020 | Quantitative proteomics |
CCLE data types
The 2012 release combined three kinds of molecular data (plus OncoMap genotyping) with drug-response data. The 2019 paper broadened the molecular layers considerably.
| Data type | 2012 (Barretina et al.) | 2019 (Ghandi et al.) |
|---|---|---|
| Mutations | Targeted massively parallel sequencing of more than 1,600 genes | Carried forward as part of the combined characterisation |
| DNA copy number | Affymetrix SNP 6.0 arrays | Carried forward as part of the combined characterisation |
| mRNA expression | Affymetrix U133 Plus 2.0 arrays | Carried forward as part of the combined characterisation |
| RNA splicing, DNA methylation, histone modification, miRNA, RPPA protein arrays | Not part of the 2012 release | Added |
| Drug sensitivity | 24 anticancer drugs, 8-point dose-response | Combined with the new data and with knockdown and knockout screens |
The 2019 paper also released a public data portal. Phase I data files are available from the Broad CCLE site itself. Whatever file you use, record the release or date you obtained it.
CCLE and DepMap
CCLE is a characterisation resource for cell lines; DepMap is the Broad cancer dependency map, which asks which genes a cancer model needs to survive. They are related efforts and are often used together, but the Broad CCLE pages do not describe DepMap, so do not treat the names as interchangeable in a methods section. For the portal, its releases and citation practice, see DepMap: Cancer Dependency Map Portal Explained. For gene-set resources used alongside expression data, see MSigDB.
Strengths and limits
Strengths
- One panel profiled on several molecular layers plus drug response, so genotype-to-sensitivity questions can be asked within one resource.
- Breadth: 36 tumour types in the original paper and more than 1,000 lines by 2019.
- Documented by named, citable papers for each phase.
Limits
- Cell lines are not patient tumours. They are propagated in culture, so findings should be treated as hypotheses for validation.
- Counts differ by paper. The 2012 paper gives 479 pharmacologically profiled lines in its abstract and 481 in the results text; cite the figure from the source you used.
- Versions differ. Data types and line numbers changed between 2012, 2019 and the phase III proteomics paper. Name the paper and the data version in your methods.
- Line identity matters. Follow-up work on a specific line, such as HCT116 or HT-29, still needs its own identity and culture checks in your lab.
Alternatives and related resources
| Resource | Use it when | Where to read more |
|---|---|---|
| DepMap portal | You need gene-dependency data on cancer models | CASRAI DepMap guide |
| cBioPortal | You want to explore genomic data through its interface; it lists a Cancer Cell Line Encyclopedia Cells entry | cbioportal.org |
| BioGRID | You want curated protein and genetic interactions for genes you found in CCLE | CASRAI BioGRID guide |
Frequently asked questions
What is CCLE?
CCLE is the Cancer Cell Line Encyclopedia, a collaboration between the Broad Institute and the Novartis Institutes for Biomedical Research that characterises human cancer cell lines genetically and pharmacologically. The first major paper (Barretina et al., Nature 2012) reported 947 genomically characterised lines.
Is the CCLE database the same as the cell line panel?
No. The panel is the set of human cancer cell lines. The database is the molecular and drug-response data generated on them. Both are called CCLE in the literature, so name the data type and the publication in your methods.
What does CCLE stand for in a methods section?
Cancer Cell Line Encyclopedia. Spell it out at first mention and cite the publication that matches the data you used, for example Barretina et al. 2012 for the original release or Ghandi et al. 2019 for the expanded characterisation.
What is the relationship between CCLE and DepMap?
They are separate names for related Broad-led efforts. The Broad CCLE site describes the encyclopedia’s own phases and publications and does not mention DepMap. The CASRAI DepMap guide covers the DepMap portal, its releases and what to cite.
Does CCLE include proteomics?
Yes, in phase III. The Broad CCLE site lists a 2020 Cell paper on quantitative proteomics as the phase III publication. The 2019 Nature paper had already added reverse-phase protein array (RPPA) data.
Can CCLE results be generalised to patient tumours?
Not directly. Cell lines are grown outside the body and accumulate changes in culture, so a cell-line finding is a hypothesis to test in patient-derived data or in vivo, not a clinical conclusion.
Sources
- Broad Institute, Cancer Cell Line Encyclopedia site: sites.broadinstitute.org/ccle (read 2026-11-07).
- Barretina J, et al. The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature 2012;483(7391):603-607. doi:10.1038/nature11003 (PMC3320027; read 2026-11-07).
- Ghandi M, et al. Next-generation characterization of the Cancer Cell Line Encyclopedia. Nature 2019;569(7757):503-508. doi:10.1038/s41586-019-1186-3 (PMC6697103; read 2026-11-07).
