TrendBio

#6 · Genomics

Assumption-light variant and repeat inference from long and short sequencing reads

3 preprints · 2 published · ↑ 1.5× vs 4-wk avg
Items per day, last 8 weekstoday: 5

The dominant thread is inference of variation, shared haplotype structure and repeat architecture directly from sequencing reads, with minimal model assumptions. SNooPy calls SNPs in long-read metagenomic data without assuming haplotype number, relationships, or divergence. bronko detects viral SNPs and minor variants through alignment-free k-mer pseudo-mapping that scales near-linearly. PBML enumerates SMEMs shared by at least k haplotypes over at least L sites on a reusable run-length PBWT index. For tandem repeats, ScatTR estimates copy number of large expansions from short reads using a maximum-likelihood Monte Carlo method, and EquiRep reconstructs repeat units from error-prone reads. What is new is explicit design for computational scale, index reuse and assumption-light inference. The repeat tools address different data types and methods rather than sharing the haplotype-index approach.

Published papers

Preprints

Topic history

DateRank
#6
#9
#9
#3
#1
#4
#5
#14
#23
#32
#21
#14
#12
#17
#17
#10
#21
#37
#27
#32
#39
#28
#35
#9

Sponsors