Fighting Cancer with Machine learning
Tell Me More

Research Focus

We are building the Cancer Dependency Map!

Precision cancer medicine

We build predictive models to predict cancer vulnerabilities from genomic profiles of tumors and cancer cell lines.

Cancer Targets Identification

We integrate functional screening and 'omics data to identify novel cancer targets as well as drugs for repurposing.


We develop computational methods and tools to facilitate the analysis of CRISPR screening in cancer models.

Small-molecule screens

We analyze highly-multiplexed small-molecule screening data from the PRISM platform to discover novel cancer therapeutic leads.

Our Team

Aviad Tsherniak

Associate Director

Philip Montgomery

Sr Principal Software Engineer

Mike Burger

Associate Computational Biologist II

Neekesh Dharia

Postdoctoral Scholar

James McFarland

Data Scientist II

Josephine Lee

Software Engineer

Josh Dempster

Data Scientist

Jordan Rossen

Associate Computational Biologist I

Allie Warren

Associate Computational Biologist I

Jérémie Kalfon

Associate Computational Biologist I

Andrew Tang

Sr. Visual Designer

Mustafa Kocak

Computational Scientist I

Phoebe Moh

Associate Software Engineer

Mariya Kazachkova

Associate Computational Biologist I

Vickie Wang

Associate Computational Biologist I

Josh Pan

Postdoctoral fellow

Yejia Chen

Software Engineer

Ashir Borah

Associate Computational Biologist I

Gwen Miller

Broad Cancer Scholar - Associate Computational Biologist

Andrew Boghossian

Associate Computational Biologist I

Nishant Jha

Software Engineer

Join Us

Therapeutic Target Discovery

Apply here

A major obstacle for treating cancer is a lack of precision medicines. Many potential targeted therapies fail to transition from preclinical models to patients due to incomplete knowledge of the drug’s mechanism of action and/or absence of robust biomarkers to identify relevant patient populations. The Target Discovery arm of the Cancer Dependency Map project aims to provide the oncology community with potential drug targets that have a high likelihood of success.

Our strategy is to find biomarkers that predict sensitivity to genetic or chemical perturbations. Investigating these relationships can lead to mechanistic understanding of the vulnerability, from which one can form therapeutic hypotheses and determine which targets are most likely to translate to patient tumors.

As part of this effort, you will work with some of the largest experimental cancer biology datasets in the world, including the Project Achilles genome-wide CRISPR screens and the PRISM Repurposing screens.

Drug repurposing analytics

Apply here

Drug repurposing offers an opportunity for rapid clinical translation of drugs that have already been optimized and proven safe in humans, circumventing a costly and slow drug discovery process. The Broad Repurposing Hub has developed the world’s foremost library of clinical drugs, and we have recently shown that high-throughput screening of these drugs across hundreds of well-characterized cancer cell lines allows for discovery of novel anti-cancer efficacy of even non-oncology drugs.

These large-scale drug screening data present many analytical challenges, and unlocking their potential requires deep integration with genomic and molecular characterization of the cancer cell lines, genome-wide measurements of their genetic vulnerabilities, as well as clinical and other datasets. As we expand the scale of this Drug Repurposing resource, we are looking for skilled data scientists, computational biologists, and machine learning practitioners to join the Cancer Data Science team and help lead data-driven discovery of new precision cancer therapeutics.

Cancer Dependency Map Genomics

Apply here

The success of cancer precision medicine requires the ability to determine optimal treatments given the molecular information encoded in each patient’s tumor. The Broad Institute’s ambitious Cancer Dependency Map initiative ( aims to create a comprehensive laboratory-based map of vulnerabilities across pediatric and adult tumors in service of facilitating biological discovery and advancing precision medicine.

We are looking for a highly motivated and talented individual to lead the Cancer Dependency Map Genomics team as part of this ambitious initiative, creating and implementing a scientific vision to take the Cancer Cell Line Encyclopedia (CCLE) effort to the next level.


Han Xu

Associate Professor, MD Anderson

Robin Meyers

Graduate Student, Genetics, Stanford University

Jared Jacobsen

Studying for AI Research

Li Wang

Computational Biologist, 10X Genomics

Jordan Bryan

Graduate Student, Statistics, Duke University

Kailash Nakagawa

Undergraduate Student, Stanford University

Quinton Wessells

Graduate Student, Biomedical Informatics, Stanford University

Remi Marenco

Bioinformation Lead, Cancer Cell Line Factory

Guillaume Kugener

Medical Student, USC

Zandra Ho

Medical Student, Brown

Andy Jones

Graduate Student, Computer Science, Princeton University

Selected publications

Contact Us

Aviad Tsherniak

Cancer Data Science
Broad Institute of MIT and Harvard
415 Main Street
Cambridge, MA 02142

Email: [first name] at