Team:Vanderbilt/Software/Stats
Demons in the Code
Mutability Score Statistics
After writing software to wrap the BioBrick API, we tried to verify the efficacy of our technique by determining how much our algorithm could optimize protein coding sequences available on the Parts Registry. We scanned over 1400 parts on the registry and determined their mutability scores, before and after optimization.
Globally Optimized
The typical optimization scheme optimizes according to a series of weights, explained in the software section. This scheme attemps to produce the lowest possible chance of mutation in most cases. On average, this scheme reduced counts as follows:
{
rateLimitingCodons: 24.489597780859917,
weightedPyrDimerCount: 17.15401567267683,
miscSites: 37.601941747572816,
hairpinSites: 20.033980582524272,
insertionSequences: 0.13661581137309292,
antiShineDelgarno: 125.69694868238558,
deaminationSites: 51.245492371705964,
repeatRuns: 26.17614424410541,
homologyRepeatCount: 0,
alkylationSites: 176.72122052704577,
methylationSites: 13.169209431345354,
oxidationSites: 149.94660194174756,
RFC10Sites: 0.13800277392510402
}