maq problem for 108bp reads.

[xiang]

I just started to use maq to map the paired-end reads of 108bp.
When I use command "maq map -e 200 -a 500 a1.bfq a2.bfq",
I always got this:

-- maq-0.7.1
[ma_load_reads] loading reads...
[ma_load_reads] set length of the first read as 108.
[ma_load_reads] set length of the second read as 108.
[ma_load_reads] 1456472*2 reads loaded.
[ma_longread2read] encoding reads... 2912944 sequences processed.
[ma_match] set the minimum insert size as 109.
[match_core] Total length of the reference: 1586004629
[match_core] round 1/3...
[match_core] making index...
[match_index_sorted] no reasonable reads are available. Exit!

I checked the data using "maq bfq2fastq a1.bfq". It shows the data is as this:

@IL41_3804:3:10:236:221/1
CAGTGATTTCGTCATTTTTCAAGTCGTTAAGTGGATATTCCTCATTTTCCATGATTTTCAGTTTTCTTGCCATATTCCATGTCCCACAGTGGACATTTCTAAATTTTC
+
CCDCDBACBABE<CCDDDCCACA;=DCCCBCACCBCCCA?BB=BCCCB@CACA<CCCC?@B@@CC?CB@<=@@0-3<6=A@@>77;5978A?<;=@?@=A?;:=8;??

I can't find any problem. Anyone has idea about this? Thanks.

[suseq]

I got the same problem with 75 bp long paired end reads. Do you find a resolution?

[nilshomer]

I just started to use maq to map the paired-end reads of 108bp.
When I use command "maq map -e 200 -a 500 a1.bfq a2.bfq",
I always got this:

-- maq-0.7.1
[ma_load_reads] loading reads...
[ma_load_reads] set length of the first read as 108.
[ma_load_reads] set length of the second read as 108.
[ma_load_reads] 1456472*2 reads loaded.
[ma_longread2read] encoding reads... 2912944 sequences processed.
[ma_match] set the minimum insert size as 109.
[match_core] Total length of the reference: 1586004629
[match_core] round 1/3...
[match_core] making index...
[match_index_sorted] no reasonable reads are available. Exit!

I checked the data using "maq bfq2fastq a1.bfq". It shows the data is as this:

@IL41_3804:3:10:236:221/1
CAGTGATTTCGTCATTTTTCAAGTCGTTAAGTGGATATTCCTCATTTTCCATGATTTTCAGTTTTCTTGCCATATTCCATGTCCCACAGTGGACATTTCTAAATTTTC
+
CCDCDBACBABE<CCDDDCCACA;=DCCCBCACCBCCCA?BB=BCCCB@CACA<CCCC?@B@@CC?CB@<=@@0-3<6=A@@>77;5978A?<;=@?@=A?;:=8;??

I can't find any problem. Anyone has idea about this? Thanks.

There is a maximum read length. You'll have to dig in the source code. Try BWA, the successor to MAQ.

[Jonathan]

I got the same problem with 75 bp long paired end reads. Do you find a resolution?

Ah, but MAQ works for 76bp PE - at least the latest version did for my dataset.
best
-Jonathan

[xiang]

I have tried more data. Now I am sure that maq can work for 108bp reads since I have already got some successful cases.

But for some data, It just reports this error. I don't know why.

[suseq]

The newest version is MAQ 0.7.1, or not? I tried to use this version but get always:
"[match_index_sorted] no reasonable reads are available. Exit! "
The rest looks similar to Xiang`s example: the reads were loaded, the reference as well, but than it stopps.

Do you have any idea why???

[stephlg]

some news?

In my lab, we sequenced 4 samples on a PE 2x60.All samples have been sequenced on the same flowcell and sequences have been generated with the same pipeline.

I used maq and I managed to align only 2 out of 4 samples. For the two non aligned samples I got the same error as you had.
I tried different parameters but I always have the same error.

I really don't understand why I couldn't align 2 samples since all samples have been processed the same way! They are in the same iFastq format.

Some news about this issue?

[lh3]

This is most likely that your reads have too many 'N's.

[xiang]

Basically, this part of data is of low quality.

What I have done for them is: chopping them down to a smaller read-length.
for example 108bp -> 93bp and then align it. You can use maq -1 93 to do this without changing the .bfq files.

[stephlg]

Actually I had a look on my data and it appeared that a lot of sequences contained a lot of Ns. So I filtered out tags with Ns and aligned them. Finally it has worked.

Thank you for your help