Relative to a reference human genome, your personal genome has about 10,000 DNA variations which affect final protein function and 3 million which do not. While “association studies” of common DNA variations with diseases yield, so far, weak predictive power and few causative mutations, researchers expect that this will be soon remedied by genome-wide sequencing. Second-generation sequencing (multiplex cycles of fluidics and imaging) has brought costs down since 2004 by
10,000-fold $300M to $30K -- and less than $1000 via targeted sequencing including coding variants (~1% of the genome), regulation, microbes and immune response (quantitated by sequencing). Polonator.org is the only of the second-generation that has open architecture for hardware, software, and wetware – and 4-fold less expensive. Similarly PersonalGenomes.org is a uniquely open effort to integrate the above genomic data with comprehensive sets of medical and non-medical traits. We are collecting over 20 terabytes of raw genomic data for each of 100,000 research subject volunteers -- which boils down to less than a gigabyte each of differences from the reference genomes and quantitative genomic and trait data. Like DNA sequencing, raw DNA synthesis has come down in cost by 7-logs since 1980, but the next challenge currently being met is applying this to “programming” organism-level functions – by developing computer-aided design, new homologous recombination instruments, lab-scale accelerated evolution and personalized stem cells.

This seminar is a joint presentation of Boston/Central New England Chapter of the IEEE Computer Society and GBC/ACM.