Author: Jakub M Bartoszewicz; Anja Seidel; Bernhard Y Renard
Title: Interpretable detection of novel human viruses from genome sequencing data Document date: 2020_1_30
ID: ac00tai9_42
Snippet: We create genome-wide phenotype analysis (GWPA) plots to analyse which parts of a viral genome are associated with the infectious phenotype. We scramble the genome into overlapping, 250bp long subsequences (pseudo-reads) without adding any sequencing noise. For the highest resolution, we use a stride of one nucleotide. We predict the infectious potential of each pseudo-read and average the obtained values at each position of the genome. Analogous.....
Document: We create genome-wide phenotype analysis (GWPA) plots to analyse which parts of a viral genome are associated with the infectious phenotype. We scramble the genome into overlapping, 250bp long subsequences (pseudo-reads) without adding any sequencing noise. For the highest resolution, we use a stride of one nucleotide. We predict the infectious potential of each pseudo-read and average the obtained values at each position of the genome. Analogously, we calculate average contributions of each nucleotide to the final prediction of the convolutional network. We visualize the resulting nucleotide-resolution maps with IGV (Thorvaldsdóttir et al., 2013) .
Search related documents:
Co phrase search for related documents- convolutional network and final prediction: 1, 2, 3, 4
- convolutional network and high resolution: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14
- convolutional network and phenotype analysis: 1
- final prediction and sequence noise: 1
Co phrase search for related documents, hyperlinks ordered by date