Origin

AI and Data for Cancer Therapeutics

Computational Biologist, Spatial Transcriptomics

$140K - $170K•0.25% - 0.50%•San Francisco, CA, US / Remote
Job type
Full-time
Role
Science, Biology
Experience
3+ years
Visa
Will sponsor
Skills
Python, R, Data Analytics, Bioinformatics, Genomics, Biology
Connect directly with founders of the best YC-funded startups.
Apply to role ›

About the role

Origin Bio is producing matched spatial transcriptomics, histopathology images and clinical datasets from real patient tumours. We’re looking for a computational biologist to own post-run QC and secondary analysis of our Xenium data, from transcript-level and cell-feature outputs through to datasets aligned with post-run H&E images.

What you’ll work on

  • Develop reproducible QC workflows for transcript detection, background signal, cell segmentation, transcript assignment, tissue and imaging artefacts, and variation across samples and runs.
  • Align post-run H&E whole-slide images to Xenium DAPI images and assess registration accuracy across each tissue section.
  • Work with pathologists to bring tumour regions and other histological annotations into the spatial data.
  • Analyse the data beyond cell typing: investigate cell states, spatial neighbourhoods, tumour–immune interactions, differential expression and other patterns that emerge from the tissue. Use relevant single-cell reference datasets where helpful.
  • Identify meaningful questions that require analysis of the data to answer. For selected questions, derive well-supported reference results and turn the work into reproducible tasks that evaluate whether AI models can reach those results from the underlying data.

Helpful past experience

  • PhD or postdoctoral research in spatial biology or equivalent hands-on industry experience. Xenium or CosMx experience is especially valuable; experience with other spatial platforms is welcome.
  • Analysis of oncology tissue, H&E whole-slide images, or multiple imaging modalities such as IHC and immunofluorescence.
  • Strong Python or R skills and experience building reproducible analysis workflows.
  • Experience with scRNA-seq or bulk RNA-seq analysis, including the use of reference data to interpret spatial measurements.
  • Sound statistical judgement: recognising noise, batch effects, experimental artefacts and confounding variables, and knowing when an apparent biological finding needs further validation.

This is a full-time role. We prefer someone who can work with us in person in San Francisco, but we’re open to the right person working remotely.

About Origin

Origin is producing matched spatial molecular, histopathology imaging, and clinical datasets using cancer tissue samples excised from real patients. We operate our own laboratories, and source tumour samples from leading biobanks.

We believe cancer is a disease ideally suited to be modelled using deep learning, however, we're largely missing the right kind of data at scale. Our initial bet is on spatial biology, helping capture the role of localisation of molecules in 3D, elucidating cell-cell signalling, differentially enriched pathways across regions, and varying cellular composition. This is particularly valuable for validating biomarkers, identifying targets and diagnostics.

As R&D increasingly relies on frontier AI systems, we need to collect real-world evidence at scale to ensure model driven findings translate in the clinic.

Origin
Founded:2025
Batch:W26
Team Size:8
Status:
Active
Location:San Francisco
Founders