I have been successfully using bwa to align my reads to indexed build36.3 human genome using the "bwa aln" command for 33 lanes of paired and data. The only options I am using are -t for multi-threading and -l 32 for the long reads up to 71bp.
However the "bwa sampe" command with all default options is not working.
I ran for a specific lane read1.sai and read2.sai with 5 million reads in the original fastq files, respectively.
Ran bwa sampe command for 36 hours after which I killed the process on my 8 quad core unix cluster with 64 Gigs of RAM memory because the .sam file was still empty, nothing written. Way too much time nothing to be written in there. Assumed something is not right with input files.
Anybody had similar experience?
I noticed that read1.sai file is larger in size than read2.sai 280Mb vs 250Mb.
I assume this means that read2 had more bad reads than read1.sai, thus there are mapped reads in read1.sai that do not have mates mapping in the read2.sai.
Could this be the problem with "bwa sampe"?
I ran each read.sai with "bwa samse" and worked great.
Thanks for the help.
However the "bwa sampe" command with all default options is not working.
I ran for a specific lane read1.sai and read2.sai with 5 million reads in the original fastq files, respectively.
Ran bwa sampe command for 36 hours after which I killed the process on my 8 quad core unix cluster with 64 Gigs of RAM memory because the .sam file was still empty, nothing written. Way too much time nothing to be written in there. Assumed something is not right with input files.
Anybody had similar experience?
I noticed that read1.sai file is larger in size than read2.sai 280Mb vs 250Mb.
I assume this means that read2 had more bad reads than read1.sai, thus there are mapped reads in read1.sai that do not have mates mapping in the read2.sai.
Could this be the problem with "bwa sampe"?
I ran each read.sai with "bwa samse" and worked great.
Thanks for the help.
Comment