Computational Scientist I - Digital Pathology Image Analysis

Broad Institute•Cambridge, MA

2d•$88,000 - $124,000•Onsite

About The Position

The Broad Institute of MIT and Harvard is seeking a highly motivated Computational Scientist I to join the Surfaceome Target Discovery team and the Getz lab. The role supports target validation by developing, optimizing, and deploying quantitative digital pathology image analysis workflows. The scientist will work with the Broad Cancer Cell Line Factory, the Dana-Farber Cancer Institute Pathology Core, and the Sellers lab in a collaborative, multidisciplinary research environment. The position focuses on whole-slide tissue images, including single-plex and multiplex immunohistochemistry (IHC) and multiplex immunofluorescence (mIF) panels. The scientist will translate target expression patterns, immune microenvironments, and spatial biology into robust, reproducible quantitative metrics for validating surface targets in patient tumor and normal adult tissue microarrays.

Requirements

Master’s or Ph.D. degree in Biomedical Engineering, Bioinformatics, Computer Science, Computational Biology, or a related quantitative life sciences discipline, with expertise in digital pathology and image analysis.
Candidates with a Master’s degree must have at least 4 years of relevant industry or academic research experience.
Hands-on experience with QuPath, including cell detection, tissue microenvironment analysis, classifier development, and custom Groovy scripting for workflow automation and scalable image analysis.
Strong understanding of tissue-based imaging assays and spatial biology technologies, including H&E, chromogenic IHC, mIF, and related imaging platforms.
Proficiency in Python, R, and/or Groovy for image analysis, spatial data analysis, statistical modeling, and workflow development.
Experience developing reproducible, well-documented computational workflows, including version-controlled code management and structured data organization practices.
Strong analytical, problem-solving, and communication skills, with the ability to work effectively in multidisciplinary research environments.

Nice To Haves

Experience with deep learning frameworks such as PyTorch or TensorFlow for tissue segmentation and classification.
Experience with commercial digital pathology platforms such as Indica Labs HALO/HALO AI, Visiopharm, or related image analysis software.
Familiarity with cloud-based computing environments such as AWS or Google Cloud Platform and containerized workflow development using Docker or similar technologies.
Experience supporting translational oncology, immuno-oncology, or biomarker discovery programs in academic, biotech, or pharmaceutical research settings.
Scientific contributions through publications, collaborative projects, or presentations involving computational pathology, spatial biology, or quantitative imaging analyses.

Responsibilities

Design, develop, and optimize machine learning and AI-based methods for digital pathology, including image segmentation, classification, spatial analysis, and predictive modeling using QuPath and other imaging platforms.
Build and maintain scalable pipelines for preprocessing, quality control, and analysis of tissue microarray and whole-slide imaging datasets from high-plex mIF and chromogenic IHC assays.
Develop and execute workflow and classifier validation strategies through benchmarking, performance assessment, and external dataset validation.
Collaborate with pathologists and scientists to assess image quality and classifier accuracy and identify imaging artifacts, staining variability, segmentation errors, and tissue integrity issues.
Partner with pathologists, assay biologists, translational researchers, and/or data scientists to turn biological and clinical questions into quantitative imaging metrics and actionable analytical outputs.
Coordinate imaging analyses and deliverables across cross-functional teams to support project goals, timelines, and study milestones.
Maintain documented, reproducible workflows, including SOP-compliant analysis records, QC documentation, version-controlled code repositories, and organized large-scale imaging datasets.