Data Scientist II

Axle•Rockville, MD

9d•Onsite

About The Position

Axle is a bioscience and information technology company that offers advancements in translational research, biomedical informatics, and data science applications to research centers and healthcare organizations nationally and abroad. With experts in biomedical science, software engineering, and program management, we focus on developing and applying research tools and techniques to empower decision-making and accelerate research discoveries. We work with some of the top research organizations and facilities in the country including multiple institutes at the National Institutes of Health (NIH). We are seeking a Data Scientist II to join our vibrant team supporting the National Cancer Institute (NCI) at the NIH in Rockville, MD. This role is embedded within NCI's Center for Biomedical Informatics and Information Technology (CBIIT), where you will directly advance cancer research by building the computational infrastructure that scientists depend on every day. You will support the full omics data lifecycle across a broad spectrum of modalities, including bulk RNA-seq, single-cell RNA-seq (scRNA-seq), spatial transcriptomics, Digital Spatial Profiling (DSP), whole genome and exome sequencing (WGS/WES), metagenomics, metabolomics, and proteomics, as well as clinical, imaging, and biospecimen data. A core part of this role involves developing workflows that integrate these modalities to support systems-level biological questions, cross-cohort studies, and NCI CBIIT initiatives. You will collaborate closely with NCI scientists, bioinformaticians, clinician-researchers, data engineers, software developers, and government stakeholders to ensure analytical infrastructure is FAIR-compliant, containerized, version-controlled, well-documented, and purpose-built for long-term reuse across the research community.

Requirements

Bachelor's degree in Data Science, Bioinformatics, Computer Science, Biological Sciences, or a related field (advanced degree preferred), or equivalent experience.
Demonstrated experience in a data-intensive role supporting biomedical research or scientific computing.
Strong proficiency in Python and R for analysis, scripting, and visualization.
Hands-on experience with at least two omics data types (e.g., bulk RNA-seq, scRNA-seq, spatial transcriptomics, proteomics, metagenomics, GWAS).
Solid understanding of statistical modeling, dimensionality reduction, clustering, differential expression, and pathway analysis.
Ability to work with structured, semi-structured, and unstructured data across relational and data lake environments.
Strong problem-solving skills with the ability to communicate effectively across technical and non-technical audiences.
Able to translate scientific needs into technical solutions and clearly articulate risks, assumptions, and limitations.
Genuine interest in biomedical and translational research.
Ability to quickly learn domain-specific terminology and workflows, with awareness of data governance, privacy, and compliance requirements for clinical and research data.

Nice To Haves

Experience building analytics solutions in platforms such as Snowflake, Databricks, or cloud data warehouses, with integrations across databases, APIs, dashboards, and application environments.
Experience with workflow and reproducibility tools used in Galaxy, Terra, Nextflow/WDL, Snakemake, Singularity, or CWL.
Familiarity with the scverse Python ecosystem (Scanpy, Squidpy, SCIMAP, AnnData) and spatial single-cell analysis methods, including PhenoGraph, Louvain/Leiden clustering, UMAP, and Ripley's L statistic, is a plus.
Experience preparing curated datasets for dashboards, APIs, and web applications.
Familiarity with Posit Connect, R/Shiny, Streamlit, Jupyter, or similar platforms is a plus.
Experience with AWS (EC2, S3, Lambda), object storage, relational databases, scheduled jobs, API integrations, and secure data movement.
Familiarity with HPC environments, SLURM/SGE, or NIH Biowulf is preferred.
Background in biomedical research, clinical research, or healthcare analytics.
Familiarity with standards such as HL7/FHIR, CDISC, or OMOP, and experience with clinical, genomic, or biospecimen data is a plus.
Experience with metadata management, data lineage, open-source code release, containerized analyses, and secure handling of de-identified or access-controlled research datasets.
Experience creating documentation, training materials, or workshops for researchers and non-coder audiences.
Ability to support tool adoption and explain workflows and results clearly is strongly preferred.

Responsibilities

Design, build, and maintain reproducible pipelines for diverse biomedical data types — including genomic, transcriptomic, single-cell, spatial, proteomic, metagenomic, metabolomic, and clinical datasets.
Develop reusable transformation logic and curated datasets supporting analytics, dashboards, APIs, notebooks, and downstream research workflows.
Support NCI CBIIT labs in their analysis workflows including bulk RNA-seq (QC, DEG, GSEA), single-cell RNA-seq (clustering, UMAP/t-SNE, cell type annotation, DEG), and Digital Spatial Profiling (annotation, QC, normalization, spatial deconvolution, volcano plots, heatmaps).
Enable reliable data movement from source systems into structured, analysis-ready formats.
Support ingestion, curation, metadata capture, source-to-target mapping, schema management, provenance tracking, and long-term maintainability of data products.
Apply statistical and ML methods — including hypothesis testing, regression, clustering, PCA, UMAP, t-SNE, and classification — to biomedical datasets.
Incorporate AI/LLM-based extraction where appropriate, with clear validation and communication to stakeholders.
Build and support interactive dashboards (Shiny, Streamlit), notebooks, reports, and APIs enabling researchers to explore multi-omics and clinical data.
Support figure generation for QC, differential expression, pathway, and spatial analyses.
Partner with data scientists, bioinformaticians, researchers, developers, and government stakeholders to translate scientific needs into technical specifications, data models, and reusable workflows that accelerate biomedical research.