VSEARCH error on amptk -filter step

[MelissaIngala]

Hello,

I am using amptk to analyze some paired-end COI diet data from bat feces. I have mostly gotten the pipeline to work, except for the amptk -filter step. My dataset included 2 mock communities, which I think amptk cannot handle, so i set my mock equal to IM3a and asked it to ignore IM3b. When I run the below command, I get the following error in the log file:

amptk filter -i cluster.otu_table.txt -f cluster.cluster.otus.fa -b IM3a -m mock_members_seqs.fas --ignore IM3b

Reading file cluster.otus.counts.fa

Fatal error: Invalid (zero) abundance annotation in FASTA file header

However, when I inspect the cluster.otus.counts.fa file, it appears to contain the abundance annotation:

>OTU1;size=1272388
AACATTATATTTTATTTTTGGTGTGTGATCCGGTATAGTAGGAACCTCTTTAAGTCTTTTAATCCGAGCCGAATTAGGCCACCCAGGTTCTCTAATTGGAGACGATCAAATTTATAATGTTATTGTAACTGCCCACGCCTTCATCATAATCTTCTTTATAGTAATGCCAATTATAATTGGAGGATTTGGAAATTGATTAGTACC
>OTU2;size=91828
AACTTTATATTTTATTTTTGGGGCTTGAGCAGGAATAATCGGAACTTCCCTAAGAATATTAATTCGAGCAGAACTAAGTCATGCCGGATCTTTAATTGGAAACGACCAAATTTATAATGTTATTGTTACTGCTCATGCATTTATTATAAttttttttATAGTAATACCAATTATAATTGGTGGATTTGGAAATTGATTAGTACC
>OTU9;size=55353

Any idea what might be causing this error? Thanks so much!