I am currently building up a inhouse-pipeline for analysis of NGS data. I apply several aligners in the pipeline including SSAHA2, BWA Mem, bowtie2 and NextGenMap. Before the mapping i split my fastq file in smaller bundles and map the bundles in parallel by the usage of MPI.
I now performed some test runs and observed an error while parsing the sam-output files. In some of the runs an alignment has an erroneous entry in the column giving the start position of the mapping on the reference genome.
Example:
anabaena_18666241 0 1_1 246"564 60 7S93M * 0 0 etc.
As you can see there is a " in the column such that when i want to cast the entry into an integer i get an error due to the " . I checked for the correct position of the alignment and saw that the " must be a 2 so the correct position is 2462564.
This error occures for NextGenMap and if i remember correctly also for one of the other aligners. The error also occurs randomly for some of the bundles and also not for one explicit read.
Does any one of you guys also observed such a fault entry in his sam output files?
Thanks for your answers.
Mchicken
I now performed some test runs and observed an error while parsing the sam-output files. In some of the runs an alignment has an erroneous entry in the column giving the start position of the mapping on the reference genome.
Example:
anabaena_18666241 0 1_1 246"564 60 7S93M * 0 0 etc.
As you can see there is a " in the column such that when i want to cast the entry into an integer i get an error due to the " . I checked for the correct position of the alignment and saw that the " must be a 2 so the correct position is 2462564.
This error occures for NextGenMap and if i remember correctly also for one of the other aligners. The error also occurs randomly for some of the bundles and also not for one explicit read.
Does any one of you guys also observed such a fault entry in his sam output files?
Thanks for your answers.
Mchicken
It vastly outperforms all others on indels.
Comment