Genetics, Harvard Medical School, Director of the Center for
Computational Genetics. 1984 Harvard PhD included the first direct
genomic sequencing method, molecular multiplexing tags, which lead to
automation & software used at Genome Therapeutics Corp. for the
first commercial genome sequence -- pathogen, Helicobacter in 1994. This multiplex solid-phase sequencing evolved into polonies (1999), ABI-SOLiD (2005) & open-source
Polonator.org (2007). Innovations in DNA reading, writing & allele replacement in
cells lead to current research & commercialization in human
genomics (Complete Genomics, PersonalGenomes.org, 23andme, Knome) , synthetic biology (SynBERC, Joule, LS9) & new ethics/security strategies.
Our ability to view and alter biology is progressing at an exponential pace -- faster even than electronics. Next generation sequencing can be used to assess inherited, environmental and epi- genomes. CLIA typically ensures reproducibility, but not necessarily highest accuracy. Genome sequence accuracy requires haplotype phase (measured, not merely inferred). Interpretation accuracy requires deep knowledge of the interactions of genomes, environments and traits. We can now move from mere correlation to causality by systematically synthesizing millions of genomic (and epi-genomic) variants via CRISPR technologies and human pluripotent stem cells. We test these technologies and interpretation software (http://GET-evidence.org
) using data and cells from the world's only fully shareable genomics resource (http://Personalgenomes.org
) -- enabling clearer preview of precision medicine. The learning objectives of this session include measuring accuracy of the underlying data, the computational integration of diverse data types -- and informed consent in an era of leaks, rich data and re-identification.