[der_eiskern]

-H multiple hits file interpretation

the file produced by using the '-H' option is gzipped ... maq docu's don't say this anywhere that I could find, unfortunately. Just name it something.gz and gunzip it ...

I unzipped it but I found the nomenclature confusing and I can't find documentation on the "-H" output organization.

Here's an truncated example of what I get when I unzip 1 of 40 .mult files:

C C comments
C R readID read_name
C B chr_name
C A readID position strand score
C E chr_name
R 0 sid:1279_11_33/1
R 2 sid:1279_11_63/1
R 4 sid:1279_11_71/1
R 6 sid:1279_11_81/1
...
R 1999958 sid:1290_1548_995/1
R 1999960 sid:1290_1548_1030/1
B ref|
A 279886 505040 - 3
A 279886 505043 - 1
...
A 1988528 9696413 - 7
A 706008 10749134 + 5
A 279886 10983194 - 4
A 279886 10983197 - 1
A 279886 10983200 - 5
A 279886 10983203 - 3
A 279886 10983206 - 5
A 279886 10983209 - 4
A 279886 10983212 - 2
A 279886 10983215 - 1
A 1134192 11511986 + 6
A 1233568 11511987 + 7
A 715394 11511996 + 6
A 1549282 12238857 - 7
A 946364 12238859 - 7
A 1627752 12971117 + 7
A 1734302 12971121 + 6
A 1193718 12997482 + 7

I expected to have more "A" lines (mapped locations) than "R" lines (reads) but maybe i misunderstood the format of what is output. It outputs all of the reads even if they didn't have multiple hits and then lists all non-unique hits with scores less than 10?

i have 999986 "R" lines and 2503 "A" lines. these 2503 mapped locations were identified by only 280 unique reads. So, as a quality control measure, does this mean that for 99.99% of my reads MAQ either failed to make any assignment to the genome or made a unique assignment?

I'm sorry if this has been answered elsewhere.

Thanks.

PS. can MAQ merge .mult (-H) files or do i have to do that myself for batched reads?