MAQ demands pairs have the same length?

[Zigster]

I'm using 0.7.1

I consistently get one of the two errors when running "maq match" when pairs have different sequence lengths (due to trimming)

maq: read.cc:61: longreads_t* ma_load_reads(void*, int, void*, int): Assertion `strncmp(name, lr->name[j], tl-1) == 0' failed.

maq: read.cc:58: longreads_t* ma_load_reads(void*, int, void*, int): Assertion `l >= size_r' failed.
Aborted

I am forced to set the -1 and -2 options to the shortest sequence in the run. This seems too restrictive to me. What is the logic behind this and why are there separate -1 and -2 options if they have to be set to the same length anyway?

[Zigster]

i suspect MAQ might be even more restrictive than I first assumed

it appears read length cannot vary at all

is this true?

[Zigster]

i guess so

# Is Maq able to deal with reads with variable lengths?

No and yes. In alignment, all the reads must be in the same length. However, you can `mapmerge' several alignments from reads in different lengths and after that all the subsequent maq commands accept variable lengths.

[dvh]

novoalign will do variable read lengths, and has all maq functionality.