#7 · Genomics
Graphs resolve haplotypes from ambiguous, mixed, and low-coverage sequencing reads
The dominant thread across these papers is casting haplotype reconstruction from sequencing reads as a graph problem. pHapCompass models read ambiguity probabilistically for polyploid phase and adds graph-based uncertainty quantification. devider applies a positional de Bruijn graph to long reads to resolve diverse viral and AMR gene haplotypes in mixtures. A pangenome-graph method uses integer programming in an alignment-free framework and is benchmarked on downsampled human MHC data at 0.1x. Other papers apply related structured inference to adjacent problems: PBML enumerates shared SMEMs on a run-length PBWT index, EquiRep reconstructs tandem repeat units via equivalence classes, and ScisTree2 uses local search for lineage trees and genotype calling. What is new is explicit handling of ambiguity and uncertainty, or of mixed and low-coverage samples, within graph-based inference.
Published papers
Long-read reconstruction of many diverse haplotypes with devider.
devider: a positional de Bruijn graph haplotyping algorithm for long-read data resolving diverse viral and AMR gene haplotypes in mixtures.
ScisTree2 enables large-scale inference of cell lineage trees and genotype calling using efficient local search.
ScisTree2 is a local-search method for inferring cell lineage trees and calling single-cell genotypes under the infinite-sites model, scaling to tens of…
Estimating the size of long tandem repeat expansions from short reads with ScatTR.
ScatTR, a maximum-likelihood Monte Carlo method estimating copy number of large tandem repeat expansions from short-read data; outperforms existing tools in…
Pangenome-based genome inference using integer programming.
Proposes an alignment-free pangenome-graph genotyping method using integer programming to infer haplotypes, benchmarked on downsampled human MHC data at 0.1x…
Accurate detection of tandem repeats from error-prone sequences with EquiRep.
EquiRep reconstructs tandem repeat units from error-prone reads via self-alignment, equivalence classes, and max-bottleneck cycle search; outperforms or…
SimGBS: a rapid method for simulating large-scale genotyping-by-sequencing data.
SimGBS simulates large-scale genotyping-by-sequencing data from reference or simulated diploid genomes, with customisable demography, pedigrees, and…
Preprints
pHapCompass: Probabilistic Assembly and Uncertainty Quantification of Polyploid Haplotype Phase
pHapCompass: probabilistic polyploid haplotype assembly modeling read ambiguity, with graph-based uncertainty quantification, genome/DNA-seq simulation…
Scalable PBWT Queries with Minimum-Length SMEMConstraints
PBML enumerates SMEMs shared by at least k haplotypes spanning at least L sites on a reusable run-length PBWT index, enabling fast targeted shared-ancestry…
Topic history
| Date | Rank |
|---|---|
| #7 | |
| #23 | |
| #20 | |
| #32 | |
| #37 | |
| #22 | |
| #23 |