NIST has a long tradition of producing highly impactful Standard Reference Materials (SRMs), which typically have a single certified value or exceptionally well characterized property with known uncertainties. New biological “’omics” measurements, particularly sequencing-based methods, can produce 103 to 109 values from biological systems and are transforming biosciences and biotechnology. Yet, the translation of technologies requires greater measurement confidence underpinned by SRMs. Data produced by the NIST-led Genome in a Bottle Consortium led to the development of the first NIST RMs in this class, with widely-used benchmark germline variant calls for seven human cell lines [1].  Artificial intelligence and machine learning hold promise to automate and improve integration of diverse ‘omics technologies, which have different strengths and weaknesses, and establish new RMs. This project would involve developing these methods, including research in deep learning for genomics and “explainable AI”, and collaborating with Genome in a Bottle Consortium members and others from companies, academia, and government.

[1] JM Zook, et al. An open resource for accurately benchmarking small variant and reference calls. Nature Biotechnology 2019, 37, 561.

key words

Artificial intelligence; Machine learning; Data science; Genomics; Sequencing; Precision medicine

Citizenship:  Open to U.S. citizens

Level:  Open to Postdoctoral applicants