Page Comparison

1st Attempt:

cat ./*_R1_001.fastq > 5RM1_R1.fastq

...

cd /project/microbiome/data_queue/seq/loc_ad1/otu

./run_slurm_mkotu.pl

Just analyzing trimmed R1s to avoid suspected merge/join bias because of 2 x 150 sequencing:

cd /project/microbiome/data_queue/seq/loc_ad1/tfmergedreads/16S/loc_ad1/trimmed

cp *.R1.fq /project/microbiome/data_queue/seq/loc_ad1/R1only/

cd /project/microbiome/data_queue/seq/loc_ad1/R1only

sed -n '1~4s/^@/>/p;2~4p' ./*.fq > ./LocAdR1.fa

vsearch --derep_fulllength $s LocAdR1.fa \
        --strand plus \
        --output $s derep.fa \
        --sizeout \
        --uc $s.derep.uc \
        --relabel $s. \
        --fasta_width 0

vsearch --cluster_unoise derep.fa --centroids zotus_vsearch.fa --sizein --sizeout

3rd Attempt:

Reran 1st attempt commands starting with “./run_slurm_mergereads.pl“

Edited “215” in the below code chunk of trim_merge.pl to “115”

system("vsearch --fastx_filter $R1tmpA --fastq_trunclen 215 --fastqout $R1tmpB --threads 32");
system("vsearch --fastx_filter $R2tmpA --fastq_trunclen 215 --fastqout $R2tmpB --threads 32");
print "vsearch step1 complete\n";
if(-e $R1tmpB && -e $R1tmpB){ 
    system("vsearch --fastq_join $R1tmpB --reverse $R2tmpB --fastaout $joinedfile --threads 32");
    unlink($R1tmpA, $R2tmpA, $R1tmpB, $R2tmpB);
}