Picrust 2: A subset of reference sequences aligned poorly to the Reference sequences

Microbial community profiling PICRUSt
1

Hi,

Im using the version of picrust2 as of May 2024.

I’m trying to validate the accuracy of picrust2 for predicting my own trait data.
I randomly sampled a subset of 16s RRNA sequences from the reference sequence fasta file (picrust2/picrust2/default_files/prokaryotic at master · picrust/picrust2 · GitHub)
/pro_ref/pro_ref.fna.

Then I removed gaps by replacing - with “”.

Then I ran place_seqs on the random subsets.

place_seqs.py -s input_seqs -o output_tree -p 1 --intermediate intermediate_folder

I got the error :
Stopping - all 4201 input sequences aligned poorly to reference sequences (–min_align option specified a minimum proportion of 0.8 aligning to reference sequences).

How can reference sequences be poorly aligned to refererence sequences?

Regards
Abel Tan

2

Hi @Abel_Tan,

I’m not exactly sure what is going on there. I would suggest making a post on the piecrust user group which the main authors actively watch.

https://groups.google.com/g/picrust-users

I think there might also be similar questions on that group around this type of analysis.

Cheers,
Jacob Nearing