The alignment length are always smaller than 10kb

[xiekunwhy]

Hi,

I use mmseqs2 to align genome sequences, but I found that the alignment length are always smaller than 10kb.

Here are the steps describle how I use mmseqs2

1. cut genome sequences into pieces, 200000bp with 10000bp overlap, and cat all pieces into one file zja.chunks.fa;
2. create database use zja.chunks.fa, mmseqs createdb zja.chunks.fa zja.chunks.db -v 2 mmseqs createindex zja.chunks.db zja.chunks.db.tmp --search-type 3
3. align each piece to database, mmseqs easy-search parts_zja200/Chr01-1-200000.fa zja.chunks.db Chr01-1-200000.m8 Chr01-1-200000.tmp -e 1e-10 -s 7.5 --min-seq-id 0.9 --filter-hits 1 mmseqs easy-search parts_zja200/Chr01-190001-390000.fa zja.chunks.db Chr01-190001-390000.m8 Chr01-190001-390000.tmp -e 1e-10 -s 7.5 --min-seq-id 0.9 --filter-hits 1 ......

and I found that, in all *.m8 files, the alignment length are always smaller than (<=) 10kb, and the results (hit number) is fewer than blastn/pblast/hs-blastn.

My question is how to unlimit max alignment length and how to set mmseqs parameters to make results as similar to blastn as possible?

Best, Kun