Author: Hayden C. Metsky; Catherine A. Freije; Tinna-Solveig F. Kosoko-Thoroddsen; Pardis C. Sabeti; Cameron Myhrvold
Title: CRISPR-based surveillance for COVID-19 using genomically-comprehensive machine learning design Document date: 2020_3_2
ID: 7tkvl894_5
Snippet: We have been developing algorithms and machine learning models for rapidly designing nucleic acid detection assays, linked in a system called ADAPT (manuscript in preparation). The designs satisfy several constraints, including on: • Comprehensiveness : Assays account for a high fraction of known sequence diversity in their species or subspecies (>97% for most assays), and are meant to be effective against variable targets. • Predicted sensit.....
Document: We have been developing algorithms and machine learning models for rapidly designing nucleic acid detection assays, linked in a system called ADAPT (manuscript in preparation). The designs satisfy several constraints, including on: • Comprehensiveness : Assays account for a high fraction of known sequence diversity in their species or subspecies (>97% for most assays), and are meant to be effective against variable targets. • Predicted sensitivity : Assays are predicted by our machine learning model to have high detection activity against the full scope of targeted genomic diversity (here, based on Lwa Cas13a activity only). • Predicted specificity : Assays have high predicted specificity to their species or subspecies, factoring in the full extent of known strain diversity, allowing them to be grouped into panels that are accurate in differentiating between related taxa.
Search related documents:
Co phrase search for related documents- detection assay and genomic diversity: 1
- detection assay and machine learning: 1, 2, 3, 4, 5, 6, 7
- detection assay and machine learning algorithm: 1, 2
- genomic diversity and machine learning: 1, 2
- high fraction and machine learning: 1, 2
Co phrase search for related documents, hyperlinks ordered by date