Merging bam files from different runs

Question

[shpak.max]

I have bam files from several runs on the same samples, e.g. MyFile_run1.bam, MyFile_run2.bam, etc. If I merge the files with

samtools merge MyFile_run* -o MyFile_Merged.bam

would there be any issues with the structure of the merged file due to different headers, indexing, etc across the initial files that would create problems with downstream analyses? I wouldn't think so, but I vaguely recall getting end of file errors when I did this with multiple runs of the same sample in the past, and don't remember what has to be done to prevent this (perhaps applying certain functions/arguments to samtools merge).

[GenoMax]

Are they all aligned to an identical reference?

[shpak.max]

As far as I know (I didn't run the alignments myself), they are aligned to the same reference genome.

[GenoMax]

Compare the headers before you merge. Otherwise you have the option of converting back to fastq and realigning.

[shpak.max]

If they were created using the same reference, the headers should match and there shouldn't be any issues with merging, correct?

However, I get header files of different length using

samtools view -H File_Runx.bam > headerx

for the different runs x.

[GenoMax]

Extracting fastq followed by realigning may be a safe bet.