No.of unique sequence decreased to 15% after chimera.uch

Commands in mothur
1

Hello ,

  1. I used greengenes reference taxonomy and alignment file. After doing Chimera.uchime and remove.seqs, I lost 85% of my unique sequences. Why ?

  2. Another question, why screen.seq did not work ? Below is the summary written.

mothur > align.seqs(fasta=stability.trim.contigs.trim.good.unique.fasta, reference=gg.refalign, flip=T)

mothur > summary.seqs(fasta=current)


Start End NBases Ambigs Polymer NumSeqs Minimum: 5 2263 370 0 3 1 2.5%-tile: 9 2266 370 0 4 312 25%-tile: 9 2266 371 0 4 3119 Median: 9 2266 372 0 5 6237 75%-tile: 9 2266 373 0 5 9355 97.5%-tile: 9 2266 376 0 6 12161 Maximum: 13 2293 376 0 8 12472 Mean: 9.01379 2266.03 372.169 0 4.71208 # of Seqs: 12472
mothur > screen.seqs(fasta=stability.trim.contigs.trim.good.unique.align, count=stability.trim.contigs.trim.good.count_table, summary=stability.trim.contigs.trim.good.unique.summary, start=9, end=2266, maxhomop=8)

It took 1 secs to screen 12472 sequences.

mothur > summary.seqs(fasta=current, count=current)

Using 8 processors.

Start End NBases Ambigs Polymer NumSeqs
Minimum: 5 2266 370 0 3 1
2.5%-tile: 9 2266 370 0 4 3339
25%-tile: 9 2266 372 0 4 33387
Median: 9 2266 372 0 5 66774
75%-tile: 9 2266 373 0 5 100161
97.5%-tile: 9 2266 376 0 6 130209
Maximum: 9 2293 376 0 8 133547
Mean: 8.99997 2266 372.232 0 4.62525

of unique seqs: 12400

total # of seqs: 133547


Looking forward for suggestion.
2
  1. I used greengenes reference taxonomy and alignment file. After doing Chimera.uchime and remove.seqs, I lost 85% of my unique sequences. Why ?

Because a lot of your unique reads were chimeras. What percentage of your total reads were discarded?

  1. Another question, why screen.seq did not work ? Below is the summary written.

Looks like it worked, what am I missing?

3

Hi Dr. Schloss,

  1. I realised from your question (what is the percentage of my total reads discarded) that I did not loose most of the total sequences. I still have 98% of total sequences. This means only 2 % of the total sequence makes 85% of rubbish unique sequences :shock: …


2) I wanted to exclude sequnce below 5 (start) and above 2266 (end). but I can still see them in summary.seq. However I can see reduction in no. of sequences.
Thank you
4

the start option removes sequences that start after the start position and the end option removes those that end before the end position.