NIST has led the Genome in a Bottle Consortium to develop widely-used benchmark variant calls for seven human cell lines [1]. GIAB has primarily used methods that align DNA sequencing reads to the reference, but long and linked read sequencing technologies are now enabling haplotype-separated “de novo assembly” of human germline and cancer genomes. These methods allow reconstruction of challenging structural changes and complex regions like the MHC that are implicated in inherited diseases and are common in cancer genomes. Human genome benchmark assemblies are essential for broad adoption of these methods, this postdoc would collaborate with a unique array of technology and informatics developers in the Genome in a Bottle Consortium to develop authoritative de novo assemblies of germline and cancer genomes. It will also be important to develop appropriate performance metrics and tools to calculate these metrics for human de novo assemblies, similar to those NIST and the Global Alliance for Genomics and Health developed for germline variants [2].

[1] JM Zook, et al, M Salit. An open resource for accurately benchmarking small variant and reference calls. Nature Biotechnology 2019, 37, 561.
[2] P Krusche, et al, JM Zook. Best practices for benchmarking germline small-variant calls in human genomes. Nature Biotechnology 2019, 37, 555.

key words

Genomics; Bioinformatics; DNA sequencing; De novo assembly; Machine learning; Reference materials; Precision medicine; Data science; Artificial intelligence

Citizenship:  Open to U.S. citizens

Level:  Open to Postdoctoral applicants