Scientist, Computational Biology

Position Overview:

We are seeking a driven, analytical Scientist, Computational Biology to join our early-stage research team. In this role, you will leverage large-scale, publicly available human biobanks to evaluate the feasibility of multi-omic disease risk prediction. You will sit at the intersection of observational epidemiology, high-throughput omics, and statistical modeling. You will mine rich human datasets to discover, evaluate, and validate multi-omic signatures that identify individuals at risk before clinical onset.

What You'll Do:

• Biobank Data Mining: Ingest, process, and perform quality control on large-scale public human datasets, with an emphasis on the UK Biobank and the All of Us Research Program.
• Multi-Omics Analysis: Analyze complex human molecular profiling data, drawing from modalities such as plasma proteomics (e.g., Olink, SomaScan), metabolomics, genome-wide DNA methylation, and transcriptomics.
• Risk Prediction & Observational Epidemiology: Apply biostatistical, observational, and machine learning methods (e.g., survival analysis, Cox models, longitudinal trajectory analysis) to evaluate phenotype associations and early disease risk.
• Feasibility & Proof-of-Concept: Design and execute fast computational experiments to determine whether specific multi-omic panels add incremental predictive value over standard clinical risk factors or polygenic risk scores.
• Cross-Functional Collaboration: Partner closely with computational scientists, assay scientists, clinical and regulatory experts, and product development stakeholders to communicate analytical findings and help prioritize targets/markers for experimental validation.

What You'll Need:

• Education & Experience:
• Ph.D. in Computational Biology, Bioinformatics, Biostatistics, Epidemiology, Human Genetics, or a related quantitative field with 0–4 years of experience (or Master’s degree with 3–6+ years of relevant experience).
• Biobank Expertise: Proven hands-on experience querying and analyzing multi-modal data in major human cohorts, specifically UK Biobank and/or All of Us.
• Omic Proficiency: Demonstrated experience analyzing at least two of the following human data types:
• Plasma proteomics
• Metabolomics
• Genome-wide DNA methylation (array or sequencing)
• Bulk/single-cell transcriptomics
• Epidemiological & Statistical Rigor: Strong background in observational study design, association testing, confounding control, and time-to-event modeling on clinical/EHR phenotypes.
• Computational Toolkit: High proficiency in Python and/or R, version control (Git), and working in cloud-based biobank environments (e.g., DNAnexus, Terra, AWS, or GCP).
• Statistical Methods & Machine Learning: Well-versed in statistical approaches applicable to biomarker discovery (e.g., high-dimensional feature selection, hypothesis testing, regularization) and experienced with standard machine learning workflows (e.g., random forests, gradient boosting, penalized regression); hands-on experience with deep learning methodologies is preferred.

Preferred / Bonus Qualifications:

• Experience constructing integrated multi-omic risk scores or combining omics with Polygenic Risk Scores (PRS).
• Familiarity with causal inference methods (e.g., Mendelian Randomization).
• Experience working in an agile, early-stage biotech startup environment.

Location, Compensation, and Benefits:

• Location: Hybrid role, 2 to 3 days onsite at our Menlo Park, CA office.
• Compensation: Annual salary range of $130,000 - $150,000, commensurate with experience. This role is also eligible for equity.