Overview In this hands-on Scientist role, you apply machine-learning methods to diverse biological data to advance TCI-ability studies in drug discovery. You will partner with computational biology, data science, and discovery teams to turn datasets into actionable insights, build reusable analysis workflows, and deliver results on a defined program timeline. The work blends protein structure, phenotypic, and multimodal data to generate testable hypotheses and impact decision-making. You will contribute to robust prototypes and clear documentation in a fast-paced, collaborative environment.
Responsibilities- Develop and apply ML and statistics for TCI-ability projects using protein structure, phenotypic, and multimodal data
- curate, harmonize, and QC structured/unstructured data; document provenance and limitations
- Build reproducible workflows for feature generation, model training, validation, and evaluation
- Integrate protein structural representations with other modalities to generate hypotheses and prioritize analyses
- Write clear, maintainable analysis code and contribute to repositories and docs
- Collaborate with cross-functional scientists to translate questions into workplans and deliverables
- Communicate methods, findings, risks, and recommendations to technical and non-technical stakeholders
- Support rapid iteration on project-specific prototypes and analyses in a dynamic discovery environment
Key requirements- Advanced degree (M.S. or Ph.D.) in computational biology, bioinformatics, computational chemistry, cheminformatics, computer science, data science, biophysics, quantitative biology, or related field
- Hands-on experience applying ML to biological, chemical, or biomedical data
- Experience with protein structure data and/or structure-derived features in research or drug discovery
- Experience integrating, analyzing, or modeling phenotypic and multimodal datasets
- Strong programming skills in Python and libraries for scientific computing and ML
- Ability to build reproducible, well-documented computational workflows and evaluate model performance
- Strong problem-solving, written communication, and cross-functional collaboration skills
- Ability to independently execute defined project work and deliver high-quality outputs on a timeline
- cross-functional collaboration
- written and verbal communication
- problem-solving
- Python
- scientific computing libraries
- machine-learning libraries
