When to combine data from two batches?

Hello! I’m sorry if this has been answered, I looked across the forum first and couldn’t find something quite the same. I submitted 2 plates for 6S V4 community profiling at the UMich microbiome core a few months apart. I (mostly) randomized my samples across the two plates and also included two samples in both plates to test for batch effects. I was wondering at one point in the mothur pipeline should I pool my data? Should I pool all the fastq files together in one folder before make.file()? Or should I process them separately until “assessing error rates”, then remove the mock groups from each and then combine them for analysis using merge.files()? Thank you so much!

Hi Erica,

It would be best to pool the samples at the make.file/make.contigs step. Doing it later - like at seq.error - will likely have the problem that the alignments have been filtered differently and won’t be compatible.

Pat

Thank you so much!!!

Hi, I tried doing this but unfortunately mothur keeps timing out during make.contigs, I think because I have so many samples (186). It runs for about 3 hours and then it just says “killed” with no error. So, I am going to do the make.contigs and some subsequent steps in 2 batches. Do you have a suggestion for where I should re-combine the files? Would a good place be, after “unique.seqs” and before “preclustering” in the MiSeq protocol? And if so, would I use merge.files()? Thank you very much!

Hmmm, that’s not really too many samples. Are you requesting enough time on the cluster? Have you looked at the mothur logfile to see if there’s a more meaningful error message? Alternatively, is there a problem with your files file?

If that doesn’t work, I’d merge the files as soon as possible like right after make.contigs using merge.files.

Pat

Okay, I figured it out and it was not a mothur issue, it was a problem with connecting to the computing core and allocating my job! When I figure out the correct thing to do I’ll post it here for future people with related issues. Thank you Pat!!

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.