Assoc. Scientist, Post Doc Fellow- Data Science for Multi-omics and Biomarker Modeling in Neuroscience

Merck Merck · Pharma · MA

Postdoctoral Research Fellow to develop and apply novel AI/ML methods linking biofluid biomarkers with tissue-based multi-omics to drive precision patient subtyping and predictive modeling in neurodegenerative diseases. The role involves designing end-to-end ML/DL pipelines and new algorithms for data ingestion, modeling, validation, and delivery of tools for in silico target perturbation and biomarker dynamics prediction.

What you'd actually do

  1. Design and implement end-to-end AI/ML pipelines for biomarker discovery and patient subgroup stratification using multi-modal data (e.g., multi-omics, clinical/lab values, imaging), including robust ETL (ingestion, QC, normalization, harmonization) and reproducible workflow orchestration.
  2. Develop and innovate on ML/DL methods tailored to study objectives: formalize hypotheses, translate mathematical ideas into implementable algorithms, conduct ablation and robustness analyses, and iterate based on emerging data and stakeholder feedback.
  3. Build and benchmark models beyond standard off-the-shelf approaches, including: multi-omics integration models at scale; models for patient subtyping and progression prediction models for in silico perturbation and biomarker dynamics.
  4. Translate model outputs into actionable insights: produce clear visualizations, summaries; deliver well-documented code, APIs, and reproducible analyses.
  5. Develop high-quality documentation and internal tools to enable reuse and scaling.

Skills

Required

  • PhD in Computer Science, Machine Learning, Artificial Intelligence, Data Science, Computational Biology/Bioinformatics, or a related quantitative field; or equivalent research experience.
  • Demonstrated experience developing and applying new AI/ML models for healthcare/biomedical data (e.g., multi-omics, clinical/lab data, real-world data, imaging).
  • Strong grounding in probability, linear algebra, optimization, and numerical methods with a deep understanding of modern ML/DL algorithms.
  • Ability to read, critique, and implement state-of-the-art ML papers, and to adapt/extend existing algorithms to novel biological problems and data types.
  • Proficiency in Python with solid object-oriented design and software engineering best practices; hands-on experience with deep learning frameworks (e.g., PyTorch) and end-to-end ML workflows; computer science foundations enabling translation from math to production code.
  • Strong scientific track record, demonstrated by peer-reviewed publications in AI/ML, computational biology, or related venues, or other scientific achievements (e.g., awards, patents, open-source contributions, grants).

Nice to have

  • Knowledge of Alzheimer’s disease or neurodegenerative biology.
  • Experience with biomarker discovery and/or patient stratification.
  • Expertise in one or more of multi-omics integration at scale (genomics, transcriptomics, proteomics; single-cell/spatial preferred), graph/network methods, causal inference.

What the JD emphasized

  • novel AI/ML methods
  • end-to-end ML/DL pipelines
  • develop and innovate on ML/DL methods
  • build and benchmark models beyond standard off-the-shelf approaches
  • peer-reviewed publications in AI/ML, computational biology, or related venues

Other signals

  • develop and apply novel AI/ML methods
  • design end-to-end ML/DL pipelines
  • develop and innovate on ML/DL methods
  • build and benchmark models beyond standard off-the-shelf approaches