The abstract is read at the publisher; the summary is JClub's.
TeraMEM detects all maximal exact matches between all pairs of hundreds-to-thousands of genome assemblies using an O(r)-space compressed index, cutting memory 7.5× and speeding up up to 2,500×.
- Why it matters: All-vs-all long exact match detection is fundamental for pangenome and population-genetics analyses, but terabase-scale panels make prior approaches too slow and memory-intensive.
- What they did: Using an O(r) compressed index built from the panel’s Burrows-Wheeler transform runs, the work computes SMEMs and MEMs in O(r) or O(r+occ) time, outputting all matches and occurrences.
- The result: On human centromere assemblies from 625 haplotypes, it uses 7.5× less memory and is 20× faster including index construction; applied to Human Pangenome Reference Consortium Release 2, it links Chromosome 2’s active -satellite HOR array to ~200 kbp inversion and ~200–300 kbp pseudopalindromic flanks.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.