Mismatches to the Reference

[DarioS]

There's no documentation I can find about how kallisto handles mismatches to the reference transcriptome. Is there any way to set a maximum value or to quantify the mismatches, such as for the purpose of allele-specific expression analysis? I notice that it can't be done with --pseudobam because it uses a placeholder CIGAR string.

If I need to do both differential transcript expression and allele-specific analysis, it seems that mapping to the genome with a traditional aligner would be necessary to be able to address both tasks and using kallisto for differential expression would simply make the methods of a journal article longer.

[pmelsted]

Mismatches to the references usually cause the k-mer matcher not to find any matches, thus it would have to use surrounding k-mers to pseudoalign the read. There are no options to specify the maximum number of mismatches and allele specific expression analysis can only be done "by hand", i.e. modifying the reference.