align.seqs starts and ends at different positions

Commands in mothur
1

hello,
I try to process 2x250 PE MiSeq data following and adapting the SOP from the mothur website. I tried to align the sequences against greengenes reference alignement, but the resutls I got do not look like the example on the web site or any example in this forum. I cannot figure out why.
Then if I screen and filter, it keeps only 10% of the sequences.

align.seqs(candidate=R2.trim.contigs.good.unique.fasta, template=/home/aline/Genomic/Granules/Granule_analysis/16_S/Mothur/amplicon_seq_2016_05/gg_13_8_99.refalign, flip=t, search=kmer, ksize=9, align=needleman, gapopen=-1, processors=2)

summary.seqs(fasta=R2.trim.contigs.good.unique.align, count=R2.trim.contigs.good.count_table)
Start End NBases Ambigs Polymer NumSeqs
Minimum: 100 716 135 0 3 1
2.5%-tile: 1629 2298 270 0 3 155778
25%-tile: 4036 5694 294 0 5 1557778
Median: 4903 6409 304 0 5 3115555
75%-tile: 5077 6777 306 0 5 4673332
97.5%-tile: 5850 6809 351 4 6 6075332
Maximum: 6361 6862 400 10 10 6231109
Mean: 4360.3 5732.32 300.502 0.405523 4.93212

of unique seqs: 1331964

total # of seqs: 6231109

2

What region are you sequencing? Could you try this with the silva reference alignment? There are really no good reasons to use the greengenes reference alignment for aligning sequences. It’s a pretty horrible alignment.

Pat