STAR Aligner Stuck at Mapping Step for Some Samples

[shadizaheri]

Hello,

We've been working on the SMaHT and dGTEx projects, which involve sequencing samples from various tissues (dGTEx includes both human and macaque samples). While using STAR v2.7.11b, we’ve run into an issue where STAR sometimes gets stuck at the mapping step without any error or warning.

Issue Details:

• The alignment process starts normally but never completes for certain samples.
• We tried breaking the original FASTQ into smaller files to isolate the problematic reads, but it doesn’t seem to be a single read causing the issue.
• The issue is inconsistent—it doesn’t happen with every sample, but it occurs often enough to be concerning (We noticed this issue specifically with some muscle tissue samples processed with TruSeq RNA.)
• The command we used for alignment and logs indicate STAR is stuck during mapping but provides no errors.
• With thousands of samples expected soon for dGTEx (human & macaque) and NHP-dGTEx, this could cause failures on a large scale.

Has anyone else experienced this, or does anyone have insights into what might be causing STAR to hang during alignment? Any suggestions would be greatly appreciated.

Thanks!