Data Scientist III, Mineral Metabolism, Levi Lab

UT Southwestern Medical Center•Dallas, TX

About The Position

Works under broad direction to provide and oversee analysis and development of algorithms and software for clinical and basic research. Oversees and participates in programs affecting data acquisition, management, and analysis in basic, translational, and clinical research, including clinical trials. Performs advanced professional work supporting biomedical and translational research, using expertise in one or more of bioinformatics, quantitative image analysis, or machine learning. Collaborates with the Principal Investigator, laboratory scientists, clinicians, and multidisciplinary partners to design rigorous analytic strategies; analyze complex biomedical datasets; and deliver reproducible analyses, publication-quality figures, manuscripts, presentations, and grant materials. Expertise in all three areas or every data type used in the laboratory is not expected. This senior data scientist independently leads analytic components of multiple research programs from study design and data acquisition through quality control, modeling, biological interpretation, validation, and dissemination. The role supports mechanistic and translational studies of injury, fibrosis, regeneration, aberrant mesenchymal cell fate, wound healing, tumor biology, and therapeutic response. It is primarily computational, with close collaboration with experimental and clinical teams to keep analyses biologically grounded and responsive to clinically important questions. The Levi Laboratory studies interactions among immune, stromal, vascular, lymphatic, neural, epithelial, and mesenchymal cell populations after injury and in disease. It integrates animal models, human biospecimens, molecular and cellular biology, quantitative imaging, clinical data, and multiomic technologies to identify mechanisms, biomarkers, and therapies that promote tissue repair.

Requirements

Required degree: Ph.D. in Computer Science, Bioinformatics, (Bio)mathematics, (Bio)statistics, Physics, Electrical Engineering, or a related field; or a Master's degree in one of these or a related field; or a Bachelor's degree with demonstrated experience in broadly defined areas of bioinformatics.
With a Ph.D., two years of post-graduation experience in data analysis and/or scientific software development.
With a Master's degree, four years of post-graduation experience in data analysis and/or scientific software development.
With a Bachelor's degree, six years of post-graduation experience in data analysis and/or scientific software development.
Advanced proficiency in R and Python for data analysis, statistics, visualization, and pipeline development, with version-control experience such as Git and reproducible analytic documentation.
Ability to independently lead complex biomedical analyses in at least one of bioinformatics, quantitative image analysis, or machine learning.
Strong command of statistical inference and experimental design, including methods relevant to the candidate’s specialty and understanding of exploratory versus confirmatory analyses.
Ability to translate biological and clinical questions into rigorous analytic plans, explain technical concepts to nontechnical collaborators, and communicate uncertainty without overinterpreting results.
Ability to evaluate emerging computational methods, benchmark performance, and implement tools with transparent documentation, testing, and quality controls.
Demonstrated ownership of complex analyses, sound scientific judgment, careful documentation, timely delivery, and effective prioritization across collaborative projects.
Excellent critical thinking, troubleshooting, scientific writing, oral communication, collaboration, and project-management skills.
Ability to provide constructive analytic leadership, mentor colleagues, improve laboratory data practices, and build productive partnerships across computational, experimental, and clinical teams.
Commitment to research integrity, reproducibility, responsible authorship, data stewardship, privacy, continuous learning, and engagement with the scientific questions underlying analyses.

Nice To Haves

Ph.D. in data science, biostatistics, bioinformatics, computational biology, computer science, biomedical engineering, or a related quantitative or biological field, with substantial experience applying advanced analytics to biomedical research.
A Master's degree in a related quantitative or biological field with significant relevant experience independently leading complex biomedical data analyses may be considered.
A Bachelor's degree in a related quantitative or biological field with extensive progressively responsible experience in biomedical data science may be considered.
Experience in relevant specialties and modalities such as single-cell or spatial transcriptomics, metabolomics, proteomics, epigenomics, CyTOF, imaging mass cytometry, multiplex tissue imaging, or quantitative micro-CT.
Experience managing and integrating large clinical, imaging, or multiomic databases with robust metadata, reproducible provenance, privacy protections, and appropriate access controls.
Experience with Linux, shell scripting, high-performance computing, containerized environments, or structured collaborative code review.
Experience developing interpretable machine-learning or statistical-learning models and evaluating performance, calibration, transportability, bias, and biological or clinical validity.
A record of peer-reviewed publications and documented analysis code or research software demonstrating meaningful contributions to computational, translational, or biomedical research.
A complete publication record, such as ORCID, Google Scholar, or PubMed, and a link to a GitHub profile, code repository, or other portfolio demonstrating reproducible analytic work, if available; a nonconfidential description of relevant analytic contributions and reproducibility practices may be provided if public links are unavailable.
Domain experience in mesenchymal progenitor and stem-cell biology, lineage tracing, aberrant cell fate, osteogenesis, chondrogenesis, fibrosis, wound repair, or extracellular matrix biology.
Experience analyzing immune-stromal, neurovascular, lymphatic, or tumor microenvironments, including macrophage states, endothelial populations, sensory and sympathetic nerves, and intercellular signaling.
Experience with computational analysis of preclinical models involving burn and musculoskeletal injury, heterotopic ossification, tendon degeneration, obesity or metabolic disease, wound healing, cancer, or metastasis; hands-on animal work is not required.
Experience in human translational studies using biospecimens, circulating rare-cell populations, biomarker discovery, diagnostic modeling, imaging, electronic health record data, or longitudinal clinical outcomes.

Responsibilities

Lead the design of data-driven research strategies with the PI, clinical collaborators, and laboratory scientists, including outcomes, sample selection, controls, replication, power considerations, data harmonization, batch mitigation, and validation plans.
Independently analyze complex biomedical datasets from raw files through biological and clinical interpretation within one or more areas of expertise: bioinformatics, quantitative image analysis, or machine learning.
For bioinformatics projects, perform and document appropriate analyses, such as quality control, normalization, batch correction, cell-type annotation, differential analysis, pathway enrichment, trajectory analysis, gene-regulatory network inference, cell-cell communication modeling, or cross-modality integration.
For image-analysis projects, develop scalable workflows for relevant imaging data, including segmentation, feature extraction, spatial quantification, and visualization as appropriate.
Integrate experimental data with clinical variables, longitudinal outcomes, public datasets, or multiomic databases where relevant to identify biomarkers, mechanisms, therapeutic targets, and clinically meaningful patient subgroups.
For statistical-modeling or machine-learning projects, develop, validate, and interpret models for disease classification, early detection, risk prediction, treatment response, or biomarker prioritization using appropriate validation and evaluation methods.
Build, document, test, and maintain reproducible analysis pipelines in R and/or Python using version control and appropriate workflow management and computing resources; use containers and high-performance or cloud computing where appropriate.
Establish and maintain laboratory standards for metadata, data dictionaries, sample tracking, file organization, quality metrics, code review, backup, controlled access, privacy, and FAIR data practices in compliance with institutional and sponsor requirements.
Partner with experimental scientists to translate computational results into validation studies and ensure data collection, assay design, and metadata support rigorous downstream analysis.
Create publication-ready figures, dashboards, tables, and data narratives for manuscripts, grants, abstracts, presentations, progress reports, and institutional reviews.
Lead computational and quantitative sections of manuscripts and grant applications; contribute to methods and results writing; maintain analytic documentation; and support responses to scientific and peer review.
Present findings at laboratory meetings and scientific forums, communicate limitations and uncertainty, and provide analytically sound recommendations.
Mentor trainees and staff in study design, statistics, computational biology, reproducible research, scientific visualization, and responsible interpretation of complex data.
Manage concurrent projects, establish priorities and milestones, identify analytic risks, and move work from exploratory analysis to validated, publication-quality completion.
Analyze and develop formalisms, algorithms, and software for biomedical research applications.
Develop and maintain software tools and infrastructure to resolve specific problems.
Oversee the development, testing, and revision of programs in a team with diverse bioinformatics expertise.
Train users and other personnel in analyzing, defining, and resolving computational problems.
Perform other duties as assigned.

Benefits

PPO medical plan, available day one at no cost for full-time employee-only coverage
100% coverage for preventive healthcare with no copay
Paid Time Off, available day one
Retirement Programs through the Teacher Retirement System of Texas (TRS)
Paid Parental Leave Benefit
Wellness programs
Tuition Reimbursement
Public Service Loan Forgiveness (PSLF) Qualified Employer
Paid holidays
On-site childcare
Healthcare benefits