Job Description

Broad Institute of Harvard and MIT, Cambridge, MA

Impact the discovery of new disease therapies at the Carpenter-Singh laboratory (in the Imaging Platform).

Join our vibrant, world-class, nonprofit biomedical research center, the Broad Institute of Harvard and MIT!

Want to devote your expertise in data science to accelerate the discovery of new medicines? Our lab develops and applies cutting-edge methods to extract quantitative information from high-throughput biological images and harness such information for drug discovery and development. We are well-known for creating CellProfiler, an open-source software used by scientists around the world to quantify biological processes in images. Cited in 19,000+ scientific papers, it has led to potential therapeutics, including several successful clinical trials in cancer.

We seek a Data Science Intern (16-week, full-time position, on-site only) to join our efforts to glean insights from large collections of images. Microscopy images contain far more information than is visible to the human eye, and our mission is to unlock that information using advanced computational approaches. By characterizing cell populations at the single-cell level, our work can transform how disease targets and therapies are discovered. 

As an intern, you will work alongside scientists and engineers in a collaborative environment and contribute to research projects aimed at advancing image-based drug discovery. Specifically, you will:

• Analyze large-scale imaging datasets using computational methods to uncover patterns in protein subcellular localization and cellular morphology.
• Apply and benchmark machine learning models to predict biological outcomes (e.g., mutation impacts or drug responses) from image-derived features.
• Evaluate and improve representation-learning approaches for single-cell image profiles.

The ultimate goal of these research projects is to develop computational strategies that:

• Predict the impact of genetic mutations from cell images
• Identify the roles of unknown genes and mutations, enabling progress toward personalized medicine
• Suggest novel uses for existing drugs to treat rare or genetic diseases

Over the past decade, we developed Cell Painting, the leading image-based profiling assay, wrote review articles about the field, created an organization (CytoData Society) to build and maintain an active community, including hosting an annual symposium and hackathon (featured in Science), and hosted the 2018 Data Science Bowl which was the largest Kaggle competition for social good in history. We launched a consortium with ten pharmaceutical companies to create the world’s largest Cell Painting dataset of >1.6 billion cells responding to over 140,000 small molecules and genetic perturbations. We demonstrated that profiling can illuminate novel connections between genes, published papers building the foundation for using deep learning for profiling, and developed software libraries for supporting this research.

Applicants must have experience in large-scale data analysis and a strong interest in biology/biomedicine. An advanced degree (Ph.D. or Master’s) in a quantitative discipline such as Computer Science, Bioinformatics, Computational Biology, Physics, or Math is highly desirable. Candidates should demonstrate outstanding personal initiative, organizational ability, effective communication skills, and the ability to work collaboratively in a team. If you are an international applicant, you are required to have a valid work permit through the end of the internship period. You do not have to actively be participating in an academic course, but you must be able to prove that you are enrolled in an institution and can confirm you will be returning to that institution during or after your internship. 

Please email Runxi Shen (contact info in the attached link) your CV and cover letter to apply for this position. In your cover letter, please describe your overall career interests, why this work would suit you, and what you would aim to learn. Please include specific examples of data analysis projects you have undertaken, including links to sample Python/R notebooks demonstrating your skills. The position starts in January 2026, with compensation determined by your educational background and experience level ($20/hr – $30/hr).