DNA methylation loss coupled to mitotic cell division promotes immune evasion of tumors with high mutation and copy number load (under review)

Methylation raw data (Idat files)
Methylation metadata worksheet

Functional mutations under positive selection in cancer predict therapeutic resistance to checkpoint blockade (under review)

Codes for implementing the convolutional neural network model to predict peptide-MHC binding
Variant call results for the exome sequencing of our cohort

Computational inference of cancer-specific vulnerabilities in clinical samples (under review)

- Source codesGitHub link to DeepDependency codes
- Regulatory networks* BayesianReal breast Shuffled breast Inverted breast Liver* ARACNeReal breast Shuffled breast Inverted breast Liver
- Sample dataCellline basal Bayesian Cellline basal ARACNe Model input Clinical_sample

Convolutional neural network model to predict causal risk factors that share complex regulatory features (under review)

GitHub link to cnnGWAS codes
DNase I profiles across 349 diverse cell types (Maurano et al)
List of cell types used in DNase I profiles
KEGG gene set from ConsensusPathDB data base
3D enhancer-promoter map based on the DHS Correlation (Maurano et al)
Data for motif scan (from JASPAR db)

Network perturbation by recurrent regulatory variants in cancer (PLOS Comput Biol 2017)

- Chromatin interactomeMCF7 HepG2 K562
- NetworksBayesian causal regulatory networkARACNe transcription network (breast cancer)ARACNe transcription network (liver cancer)PCAPMI transcription network (breast cancer)PCAPMI transcription network (liver cancer)Integrated physical protein interaction networkProbabilistic functional protein association network

Global transcription network incorporating distal regulator binding reveals selective cooperation of cancer drivers and risk genes (Nucleic Acids Res 2015)

Global network
Prior framework