Research Scientist/Engineer 1

University of Washington Medical Center•Seattle, WA

4d•$65,004 - $70,008

About The Position

Field of Research: Computational biology/bioinformatics with emphasis on spatial transcriptomics, proteomics, and immunosequencing using Pixel‑seq and an immune receptor‑focused multimodal expansion (“ImmunoPixel‑seq”). Work includes NGS data processing, spatial barcode mapping, single‑cell & spatial analysis, and cell segmentation in brain, tumor, and other tissues. Purpose of the Research project(s): To build and operate robust, reproducible pipelines that transform raw sequencing + barcode map into high‑quality, spatially resolved single‑cell datasets; to deliver disease‑focused analyses that advance mechanistic discovery and translational hypotheses (e.g., tumor microenvironment, neuroanatomical circuits).

Requirements

Programming & data: Python (numpy/pandas), basic R (Seurat/tidyverse), bash; Git; Linux.
NGS data processing: BCL→FASTQ demultiplexing; adapter/quality trimming; UMI handling; QC with MultiQC; alignment/quantification to reference.
Spatial omics: Pixel‑seq barcode→(x,y) mapping concepts; creation of spatially annotated objects (AnnData/Seurat).
Segmentation: Practical use of Cellpose/FICTURE (or similar); basic image QC.
Single‑cell & spatial analysis: Normalization, clustering, label transfer; spatial neighborhood/domain analyses (e.g., with Squidpy/Giotto).
Reproducibility & automation: Snakemake or Nextflow; containerization (Apptainer/Docker); clean documentation; basic SLURM job submission.
Communication: Clear writing of READMEs, short analysis memos, and figure captions for collaboration with biologists/clinicians.
Linux/HPC usage; Slurm job submission, resource requests, and environment management.

Nice To Haves

Probabilistic modeling: scVI/scANVI/totalVI for RNA and RNA+protein integration.
GPU experience: PyTorch/CUDA for segmentation/model inference.
Data stewardship: DVC or equivalent data versioning; basic dashboarding/monitoring (Prometheus/Grafana).
Domain breadth: Prior coursework/research in biochemistry or genetics; interest in medical/MD‑PhD pathways.
DevOps‑lite: GitHub Actions CI, environment pinning, reproducible reference bundles, and runbooks.
Experience assisting with server upgrades in collaboration with IT (CUDA/cuDNN & GPU driver stacks, Slurm client updates, module systems).
Basic familiarity with configuration/monitoring for research workflows (e.g., Ansible basics, Prometheus/Grafana dashboards) under IT guidance.
Storage and I/O awareness for high‑throughput data (scratch NVMe vs. bulk); performance troubleshooting for pipelines.

Responsibilities

End‑to‑end data processing (BCL/FASTQ → QC → counts) – 20%
Demultiplexing, adapter/quality trimming, UMI handling, alignment/quantification; generation of MultiQC reports and run manifests.
Spatial barcode mapping & registration – 15%
Build/validate barcode→(x,y) maps for Pixel‑seq; error correction; join gene/protein counts to spatial coordinates; QA of mapping rates.
Segmentation & QC – 20%
Apply/benchmark nuclei or whole‑cell segmentation (e.g., Cellpose/StarDist/SAM); maintain curated masks and QC thumbnails.
Downstream single‑cell & spatial analysis – 20%
Create annotated data objects (e.g., AnnData/Seurat); normalization, clustering, label transfer; spatial neighborhood/domain analysis; multi‑omic modeling for RNA+protein where applicable.
Pipeline automation & reproducibility – 10%
Implement/maintain Snakemake/Nextflow workflows with containers (Apptainer/Docker), CI tests, and clear documentation.
Project support, collaboration & reporting – 7%
Prepare figures/tables; concise analysis memos; contribute to methods sections.
Light server/environment maintenance & upgrades (DevOps‑lite) — 5%
Build and update containerized analysis environments, maintain conda/uv environments.
DevOps‑lite & data stewardship – 3%
Maintain analysis environments/containers; basic SLURM job scripts; coordinate with IT on storage/backup hygiene.