Explore the Cancer Dependency Map

DepMap is excited to announce several important updates and pipeline improvements in this release. This release has been focused on changes to our processing pipelines and the format in which we release data. See this forum post for more information. Due to the large number of significant changes in this release, if you have previously used DepMap data, we strongly recommend consulting the README for this dataset before downloading the newest release.

Data Release Schedule Update After this quarter, DepMap will begin releasing new data once every 6 months, as opposed to every quarter. We will continue to announce new portal features, tools, and other resources as they are released throughout the year. However, new DepMap data will only be released in Q2 and Q4 of each year. Our next data release will be in 22Q4 (November 2022).

Ancestry Bias Correction We have found that cell-line specific alterations in the targeted sequences of sgRNAs can preclude Cas9-mediated genome editing, introducing ancestry-related biases into the Achilles data. To account for this, we will remove data from guides with SNV-induced sequence mismatches in a given cell line (~1.5% of sgRNAs on average). This correction allows Chronos to compute gene effects without being biased by the reduced cutting efficacy produced by these sgRNA-variant interactions. Please stay tuned for a preprint describing these analyses in more detail. In the meantime, check out the related web tool for more information.

New Terra Workspace for CCLE BAM Files To facilitate easier access to the publicly available sequencing data from the CCLE project, we have created a Terra workspace to share these data. This also includes germline variant calls using HaplotypeCaller. For further information please see our post on the DepMap forum here.

Updates to Cell Line Metadata We continue to update and improve our cell line metadata which can be found in the sample_info file on the data downloads page. To more clearly annotate information about engineered cell lines we added the columns model_manipulation and model_manipulation_details, and correspondingly removed ‘engineered’ as a category in the lineage and primary disease columns so these represent the actual values. To document cell lines that come from the same patient or that are derivatives, we added patient_id and parental_depmap_id columns. Our patient IDs indicate known isogenic relationships based on SNP-fingerprinting and information reported in the Cellosaurus and may not include all relationships. Finally, we have added Cellosaurus_NCIt_disease and Cellosaurus_issues columns by pulling this information from the Cellosaurus (v41), and have updated our age and sex annotations for a large number of cell lines by gathering this data from the Cellosaurus and other sources.

Misidentified Cell Lines This quarter we reviewed all of the ICLAC (v11) misidentified cell lines in DepMap, in most cases verifying that DepMap has the correct sample. In cases where no known authentic stock exists (such as KPL1 which is actually MCF7), we have noted the misidentification in the depmap_public_comments column of the sample_info file and have updated the cell line metadata accordingly. For example, the age, sex, lineage, and other metadata for KPL1 has been updated to be the same as MCF7 and it has been marked as an MCF7 derivative. More details can be found in this forum post.

Upgraded RSEM and STAR We updated STAR to v2.7.10a, RSEM to v1.3.3, and GENCODE to version 38 (used for generating index for STAR and reference for RSEM) in our expression pipeline.

SNP Fingerprinting Changes Using SNP-fingerprinting, we have identified four cell lines where the stock used by DepMap is not from the expected cell line. For three of these cell lines, HSB2 (ACH-001737), COLO699 (ACH-001041), and NCIH1339 (ACH-000921) we have decided to keep the data, using the same DepMap ID, but rename the cell line and update the metadata to annotate these as derivatives of cell line the data is from. For the fourth, HS571T (ACH-001092), we have decided to remove our data, which includes mutations from hybrid capture data, and copy number from SNP array data. More details can be found in this forum post. We apologize for any impacts these errors might have had.