Chimera checking, analysis of Sanger sequencing data in qiime

[Masanobu Hiraoka]

Hello, everyone,

I would like to analyze microbiome at human upper respiratory way with qiime pipeline, especially ANCOM that was read by Sanger sequencing(clone library method).

To make data more reliable, I am thinking of excluding chimera sequences.

When I do chimera checking, what and when is the best way? ChimeraSlayer, uchime or usearch? Before clustering(qiime1) or after clustering(qiime2)?

I was suggested some processing in this Forum before(post split fasta file import), and succeeded in converting data into qiime2 pipeline.

After I imported data to qiime2, I did the clustering(q2-vsearch) at 99.6% identity, exclude singleton, mitochondria and chloroplast. And further statistical analyses were performed with rarefied artifacts(commands are not shown).

Thank you

//Masanobu

commands;

Qiime1;

add_qiime_labels.py -i fasta_dir -m example_mapping4.txt -c InputFileName -o combined_fasta

validate_mapping_file.py -m example_mapping4.txt -o validate_mapping_file_output  

Qiime2;

qiime tools import \

 --input-path Raw_Files/combined_seqs.fna \

 --output-path Raw_Files/seqs.qza \

 --type 'SampleData[Sequences]'

qiime vsearch dereplicate-sequences \

  --i-sequences Raw_Files/seqs.qza \

  --o-dereplicated-table Raw_Files/table.qza \

  --o-dereplicated-sequences Raw_Files/rep-seqs.qza

qiime vsearch cluster-features-de-novo \

  --i-table Raw_Files/table.qza \

  --i-sequences Raw_Files/rep-seqs.qza \

  --p-perc-identity 0.996 \

  --o-clustered-table Raw_Files/table-dn-996.qza \

  --o-clustered-sequences Raw_Files/rep-seqs-dn-996.qza

qiime phylogeny align-to-tree-mafft-fasttree \

  --i-sequences Raw_Files/rep-seqs-dn-996.qza \

  --o-alignment Raw_Files/aligned-rep-seqs.qza \

  --o-masked-alignment Raw_Files/masked-aligned-rep-seqs.qza \

  --o-tree Raw_Files/unrooted-tree.qza \

  --o-rooted-tree Raw_Files/rooted-tree.qza

#Removing reads that don't appear in at least two samples

qiime feature-table filter-features \

  --i-table Raw_Files/table-dn-996.qza \

  --p-min-samples 2 \

  --o-filtered-table Results_Files/table-dn-996.qza

#Removing the mitochondria and chloroplasts;silva138.1full

qiime taxa filter-table \

  --i-table Raw_Files/table-dn-996.qza  \

  --i-taxonomy Raw_Files/full_taxonomy.qza  \

  --p-exclude mitochondria,chloroplast  \

  --o-filtered-table Results_Files/full_table-dn-996_Removed.qza

  

ref;

This is my history of commands list.

 "Diversity core metrics of Sanger sequencing data: " in qiime2 forum.

[TonyWalters]

Hello,

Although I haven't looked at the benchmarks recently, Chimeraslayer hasn't been updated in quite some time, so I would go with the vsearch option that's built into QIIME2.

See this tutorial:
https://docs.qiime2.org/2021.8/tutorials/chimera/

It looks like you would be able to do the chimera checking after your  qiime vsearch cluster-features-de-novo command (and you'd want to modify the commands after this to use the chimera filtered sequence artifact file).

-Tony

[Masanobu Hiraoka]

Thank you for replying!

I missed the tutorial of chimera checking in qiime2!

OK, I will do the command "qiime vsearch uchime-denovo" after clustering.

If I have any questions, I'll ask them on qiime2.

// Masanobu