Somatic Structural Variation Discovery in Multiple Cancer Genomes

on YouTube ↗10m

" summary

A pipeline called LAB-C takes multiple whole-genome BAMs plus targeted deep sequencing of ~1,000 somatic SNVs, builds a tumor lineage with LICHeE, clusters discordant read-pair breakpoints (using a method like commonlaw), and places SV clusters on internal or leaf nodes under a parsimony score that favors ancestral placements. Results show more somatic SV calls and lower false negatives than single-sample calling, including placing ~60 lesion clusters restricted to three samples and detecting a clear somatic inversion at the IG locus.

AI-written summary · about

" chapters (15)

1. 0:00Intro
2. 0:22Structural Variations (SVS) with phenotypic effect
3. 0:49Paired-end Sequencing and SV
4. 1:51Input Data
5. 2:50LAPSI's main idea
6. 3:11LAPSI's assumptions
7. 3:24LAPSI's workflow 0. Reconstructing the lineage tree, using the VAFS of somatic SNVs across samples.
8. 3:470. Reconstructing the lineage tree (Popic et al. Genome Biology 2015)
9. 4:00Breast Cancer Lineage Trees
10. 4:141. Clustering discordant paired-end reads
11. 4:37Maximum Valid SV Cluster Matrix Let n be the number of samples and m be the total number of maximal valid clusters
12. 5:08LAPSI's basic objective function
13. 5:28A greedy algorithm
14. 6:59Sub-clonal deletions and a somatic inversion
15. 7:44Example: view of a somatic inversion on IGV