Data Scientist Fellow, Bioinformatics and Image Analysis

Image analysis pipelines. Develop and maintain automated pipelines for Cytation 9 images, including quality control, illumination and background correction, segmentation, tracking, and feature extraction for cultures such as iPSC colonies and, later, organoids.

Quantitative readouts for the loop. Define and validate metrics that describe culture state and growth (for example confluence, colony morphology, differentiation markers, and viability), with uncertainty estimates, so that they can be used as decision inputs by the autonomous system.

Closing the loop. Work with the Laboratory Automation Engineer to deliver analysis results back to the workflow orchestration layer in a machine-readable form and on a timescale the loop can use. Define the interfaces, the failure behavior, and the checks that stop bad data from driving a bad decision.

Environmental data integration. Collect, clean, and align environmental monitoring data (for example temperature, CO2, O2, humidity, and other sensor streams) with imaging results. Analyze how conditions relate to culture outcomes and flag drift, excursions, and sensor faults.

Statistical and machine learning methods. Apply and develop statistical, machine learning, and deep learning methods for image and multimodal data, with attention to validation, reproducibility, and batch effects. Use the existing Cytation and instrument software where it helps, and build custom methods where it does not.

Bioinformatics. Integrate downstream omics data, including single-cell data, generated from the experiments where they are available, and link them to image-derived phenotypes.

Data management. Work with IDSS data engineers so that images, derived measurements, and metadata are stored, versioned, and findable in line with FAIR principles and NIH data management and sharing requirements.

Research and communication. Publish methods and results, present to NIAID and collaborators, document pipelines and standard operating procedures, and train staff to use the analysis tools.

Education. Doctoral degree (PhD or equivalent) in bioinformatics, computational biology, biomedical engineering, computer science, physics, or a related quantitative field, normally completed within the last five years.

Image analysis. Demonstrated experience analyzing microscopy or other biological image data, including segmentation and quantitative feature extraction.

Programming. Strong Python, including scientific and imaging libraries (NumPy, SciPy, pandas, scikit-image, or similar) and at least one machine learning framework (PyTorch, TensorFlow, or similar).

Statistics and experimental design. Strong grounding in statistics and experimental design, and experience validating analysis methods.

Reproducible practice. Experience with Git and reproducible analysis practices, and with working in a Linux or high-performance computing environment.

Research record. A record of research output, such as peer-reviewed publications, preprints, or released software.

Collaboration. Ability to work in a close, iterative partnership with an automation engineer and biologists, and to explain methods and results clearly to both.

Clearance. Ability to obtain and maintain NIH/HHS security clearance and meet federal contractor suitability requirements.

Experience with high-content or live-cell imaging systems, especially Agilent BioTek Cytation or Gen5 software.

Experience with image analysis tools such as CellProfiler, Cellpose, StarDist, napari, or ImageJ/Fiji, and with deep learning methods for segmentation and classification.

Experience with stem cell, iPSC, organoid, or tissue culture systems and their imaging.

Experience with time-series or multimodal analysis, including sensor and environmental monitoring data.

Experience with self-driving laboratories, active learning, or Bayesian optimization.

Experience with single-cell omics analysis (scRNA-seq) and with workflow tools such as Nextflow or Snakemake.

Familiarity with laboratory orchestration software such as MADSci, and with Docker or Kubernetes.

Experience with high-performance computing environments (for example NIH Biowulf).