Back

Multi-context seeds enable fast and high-accuracy read mapping

Tolstoganov, I.; Martin, M.; Sahlin, K.

2025-02-18 bioinformatics
10.1101/2024.10.29.620855 bioRxiv
Show abstract

A key step in sequence similarity search is to identify shared seeds between a query and a reference sequence. A well-known tradeoff is that longer seeds offer fast searches but reduce sensitivity in variable regions. We introduce multi-context seeds (MCS), which allow to store seeds with different lengths in the same index structure, thus retaining the advantages of both short and long seeds. We demonstrate the applicability of MCS by implementing them in strobealign. Strobealign with MCS substantially improves accuracy compared to the previous version with little cost in runtime and no memory overhead.

Published in Genome Biology (predicted rank #1) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

Genome Biology · published here
○ 637 papers in training set
Top 0.1%
29.9%
Similar papers in this journal
2
Bioinformatics
◐ 1204 papers in training set
Top 2%
11.5%
Similar papers in this journal
3
Nature Communications
○ 5641 papers in training set
Top 20%
8.6%
Similar papers in this journal
4
Genome Research
● 468 papers in training set
Top 0.9%
6.5%
Similar papers in this journal
50% of probability mass above
5
Nature Methods
○ 385 papers in training set
Top 2%
5.3%
Similar papers in this journal
6
Nucleic Acids Research
◐ 1281 papers in training set
Top 5%
3.9%
7
Bioinformatics Advances
◐ 203 papers in training set
Top 2%
3.9%
8
GigaScience
◐ 212 papers in training set
Top 1.0%
3.9%
9
NAR Genomics and Bioinformatics
◐ 242 papers in training set
Top 1%
3.3%
10
BMC Bioinformatics
○ 457 papers in training set
Top 2%
3.3%
11
PLOS Computational Biology
● 1863 papers in training set
Top 11%
3.1%
12
Nature Biotechnology
○ 172 papers in training set
Top 2%
2.5%
13
Briefings in Bioinformatics
◐ 354 papers in training set
Top 4%
2.0%
14
Nature Computational Science
○ 55 papers in training set
Top 0.8%
1.6%
15
iScience
○ 1154 papers in training set
Top 29%
1.1%
16
Molecular Biology and Evolution
◐ 542 papers in training set
Top 4%
1.1%
17
Algorithms for Molecular Biology
○ 17 papers in training set
Top 0.2%
1.1%
18
Cell Systems
○ 201 papers in training set
Top 5%
0.8%
19
Scientific Reports
○ 3612 papers in training set
Top 79%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.