By downloading this data you agree to our Terms and Conditions.
DepMap currently stores raw omics data from publicly-releasible cell line models generated from the Cancer Cell Line Encyclopedia (CCLE) on the Registry of Open Data on AWS. See our page on the registry for a description of the data set. A data dictionary depmap-omics-ccle-datadict.json is provided in this repo. Currently included are WES/WGS/RNA CRAM/BAM files for ~1000 cancer cell lines and unfiltered mutation calls (VCF) for WGS samples.
Since this data is publicly readable, aws s3 commands should include --no-sign-request, e.g. aws s3 ls 's3://depmap-omics-ccle' --no-sign-request.
See s3://depmap-omics-ccle/docs/data_dictionary.json.
An inventory of the objects in the data folder is available in the s3://depmap-omics-ccle/metadata folder in several formats:
aws s3 ls 's3://depmap-omics-ccle/metadata' --recursive --no-sign-request2025-09-29 14:31:28 878423 metadata/inventory.csv
2025-09-29 14:31:30 2955629 metadata/inventory.json
2025-09-29 14:31:30 251731 metadata/inventory.parquet
Objects referenced in the inventory are organized by data type (rna/wes/wgs), then file type, e.g. s3://depmap-omics-ccle/data/wgs/cram/CDS-051xn7.cram. Most, but not all, CRAM/BAM files have indexes available.
AWS HealthOmics can be used to analyze the raw data (i.e. CRAM/BAM files) on AWS without manually provisioning all the underlying infrastructure. This includes Ready2Run workflows like NVIDIA Parabricks for WGS preprocessing, somatic variant calling, RNA alignment, and more.
Workflows used by the DepMap portal are actively maintained in the case of WGS and short read RNA. These are stored on GitHub:
Each workflow is written in WDL and is accompanied by a JSON file with default workflow input values, e.g. workflows/align_sr_rna/align_sr_rna.wdl and workflows/align_sr_rna/align_sr_rna_inputs.json, which can be used as HealthOmics parameter template files.
Some tasks use public Docker images on DockerHub, but most are custom images built by us. Their Dockerfiles can be found either alongside the WDL (e.g. workflows/quantify_sr_rna/salmon.Dockerfile) or in the workflows/common folder if the image is shared by multiple workflows (e.g. workflows/common/star_arriba.Dockerfile).
For latency reasons (and because our container registry is private and on Google Cloud) we recommend building your own images from these Dockerfiles and pushing them to Elastic Container Registry in the AWS region where you run HealthOmics workflows. Then, the docker_image and docker_image_hash_or_tag task inputs can be overridden to refer to your image URLs in ECR.
See the HealthOmics container images docs for more information on where your workflow's images can be stored.
We also recommend the GATK Best Practices site as a reputable source of WDL workflows.
In addtion to WDL, you can run your own private HealthOmics workflows written in CWL or Nextflow.
Running Nextflow workflows is also supported in AWS with AWS Batch as the execution backend. See the Nextflow AWS docs for instructions or read Seqera's tutorial. Verified Nextflow workflows for analyzing omics data can be found at nf-core.