Bioinformatics Scientist - Gene Regulation & Transformer Modeling

Role overview As a Bioinformatics Scientist specializing in genomic foundation models, you will lead computational analysis of multi-modal genomics data (scRNA-seq, ATAC-seq) to identify transcription factor combinations driving desired cell state conversion. This role bridges classical gene regulatory network inference and modern foundation model approaches, requiring deep expertise in single-cell genomics analysis, transcriptional regulation biology, and demonstrated interest in applying transformer-based methods to cellular reprogramming problems. You will work on traditional multi-platform genomics analysis and on integrating and fine-tuning foundation models, collaborating closely with wet lab teams to translate computational predictions into experimental designs for our cell fate engineering platform.

Primary responsibilities

Perform genomics analysis

across scRNA-seq and ATAC-seq data from human, NHP, and mouse gametogenesis to identify transcription factors governing cell state trajectory

Integrate classical and modern approaches

by combining GRN inference methods with transformer-based models to create hybrid TF ranking systems, leveraging both motif-guided statistical learning and self-supervised deep learning representations

Build version-controlled data platforms

harmonizing gametogenesis datasets across multiple modalities and sequencing platforms, preparing integrated datasets for both traditional analysis and foundation model fine-tuning

Develop trajectory inference workflows

using RNA velocity, pseudotime analysis, and optimal transport models to map cellular transitions, identifying critical commitment points where TF interventions are most effective

Apply chromatin accessibility analysis

and TF motif enrichment to decode regulatory grammar at key transition junctions, identifying synergistic TF combinations and repressive factors required for cell fate conversion

Collaborate with experimental teams

on screen design and interpretation, translating computational predictions into biologically interpretable experimental plans specifying which TF combinations to test and validation strategies

Establish automated feedback loops

for ingestion of screening NGS data, implementing active learning strategies that prioritize informative experiments and building retraining pipelines that continuously improve prediction accuracy

Build computational infrastructure

for reproducible bioinformatics workflows and foundation model fine-tuning

Required qualifications

PhD in Bioinformatics, Computational Biology, or related quantitative field

(or MS with 5+ years relevant industry experience)

Multi-platform single-cell RNA-seq expertise:

hands-on analysis of data from at least two different platforms, including platform-specific troubleshooting and quality control

Multi-modal genomics proficiency:

experience with ChIP-seq, CUT&RUN, or ATAC-seq analysis including peak calling, differential accessibility, and TF motif enrichment

Computational TF identification background:

applied computational methods (GRN inference, chromatin accessibility, perturbation screens, ML) to identify transcription factors for cell fate conversion beyond literature-curated lists

Cellular reprogramming knowledge:

prior research experience (computational or experimental) in cell fate conversion, direct reprogramming, transdifferentiation, iPSCs, or differentiation systems with deep understanding of transcriptional regulation

Strong programming skills:

Python and R with proficiency in Scanpy/Seurat, standard genomics toolkits, and statistical analysis for high-dimensional data

Foundation model interest:

familiarity with transformer architectures in genomics through coursework/self-study/application, OR strong demonstrated interest with clear learning approach

Strong publication record and demonstrated cross-functional collaboration

with experimental biologists

Preferred qualifications

Experience fine-tuning or applying genomic foundation models with demonstrable results, or contributions to bioinformatics tools incorporating transformer architectures

Prior computational research in cellular reprogramming; extensive experience with multiple GRN inference methods and perturbation response modeling

Background in trajectory inference beyond basic pseudotime, GRN for biological networks, or Bayesian approaches to genomics

GPU cluster experience for model training, multi-omics integration methods, and cross-species genomics analysis

What we offer

On-site work in the US Pacific Northwest in state-of-the-art facility

Unique opportunity at early-stage biotech startup

where you'll shape computational strategy from the beginning, build infrastructure from scratch, and have direct impact on our computational approaches for years to come

Autonomy and ownership

to design analysis frameworks, choose methodologies, and pioneer novel approaches without bureaucratic constraints, with competitive compensation including equity participation

Mission-driven impact

developing novel technologies for fertility medicine

Competitive compensation, equity participation, and comprehensive benefits

#J-18808-Ljbffr