chimeric vs unaligned reads

Hello,

I’m trying to obtain the chemic alignments from a BAM file that originally was generated by STAR and that only has a list of unaligned reads at the end of the file (so no SA tag).

If I convert that BAM file back into a fastq file with samtools and then rerun STAR with –chimOutType WithinBAM, will I label some of these unmapped reads as chimeric ones?

Is there a risk of loosing information/reads by doing so? Unfortunately I do not have the original fastq files and I’m trying to find a way around.

Thank you!

Best,
Andrea