Hey all,
I have read many suggestions about merging two or multiple BAM files using samtools merge function. I have also read in the documentation how it works. However, none of the solutions or suggestions seem to work in my case.
From the beginning, I had reads which I align using SMALT and NextGenMap (individually), obtain the alignment in SAM, convert to BAM, sort and do other calculation. I was using these BAMs to generate consensus sequence using mpileup to be used in my annotation pipeline.
Now, I want to merge BAMs from two alignment methods namely SMALT and NextGenMap, and use this merged BAM to generate consensus sequence to be used in my annotation pipeline. I want to do so because I believe that there will be many reads aligning to reference at the same configuration in common and that it will also make my further predictions more accurate. While I use samtools merge, on my sorted bam files, it generates a BAM file which does seem to satisfy the statistics but after generating consensus sequence from this bam, there are about only 8-10% of the contigs in this new consensus sequence file as compared to consensus sequence files generated by BAMs of SMALT and NextGenMap. This seems to miss out on many predicted results I obtain individually.
Anyone has any idea what I can do in such situation or may be in dept of how samtools merge actually puts everything together. Many thanks!!
I have read many suggestions about merging two or multiple BAM files using samtools merge function. I have also read in the documentation how it works. However, none of the solutions or suggestions seem to work in my case.
From the beginning, I had reads which I align using SMALT and NextGenMap (individually), obtain the alignment in SAM, convert to BAM, sort and do other calculation. I was using these BAMs to generate consensus sequence using mpileup to be used in my annotation pipeline.
Now, I want to merge BAMs from two alignment methods namely SMALT and NextGenMap, and use this merged BAM to generate consensus sequence to be used in my annotation pipeline. I want to do so because I believe that there will be many reads aligning to reference at the same configuration in common and that it will also make my further predictions more accurate. While I use samtools merge, on my sorted bam files, it generates a BAM file which does seem to satisfy the statistics but after generating consensus sequence from this bam, there are about only 8-10% of the contigs in this new consensus sequence file as compared to consensus sequence files generated by BAMs of SMALT and NextGenMap. This seems to miss out on many predicted results I obtain individually.
Anyone has any idea what I can do in such situation or may be in dept of how samtools merge actually puts everything together. Many thanks!!
