denoise-paired error

[idupanloup (Isabelle Dupanloup)]

I’m using qiime2-2019.4.

And when i run the following command:
qiime dada2 denoise-paired --i-demultiplexed-seqs demux-paired-end.qza --p-trim-left-f 0 --p-trim-left-r 0 --p-trunc-len-f 40 --p-trunc-len-r 80 --o-representative-sequences rep-seqs-dada2.qza --o-table table-dada2.qza --output-dir denoising --p-n-threads 24
I get this error:
Remove chimeras (method = consensus)
Error in isBimeraDenovoTable(unqs[[i]], …, verbose = verbose) :
Input must be a valid sequence table.
Calls: removeBimeraDenovo -> isBimeraDenovoTable
Execution halted

The import in qiime looks OK and was done with the following commands (and outputs):
qiime tools import --type ‘SampleData[PairedEndSequencesWithQuality]’ --input-path fastq_files --input-format CasavaOneEightSingleLanePerSampleDirFmt --output-path demux-paired-end.qza
Imported fastq_files as CasavaOneEightSingleLanePerSampleDirFmt to demux-paired-end.qza

qiime demux summarize --i-data demux-paired-end.qza --o-visualization demux-paired-end.qzv
Saved Visualization to: demux-paired-end.qzv

any idea of what could be the problem ?
many thanks, Isa

[Nicholas_Bokulich]

Hi @idupanloup,
This is a pretty common technical issue that has been discussed many times on the forum: Search results for 'Calls: removeBimeraDenovo -> isBimeraDenovoTable order:latest' - QIIME 2 Forum

E.g., see here for an explanation:

dada2 denoise-paired error using qiime 2019.4

Hi @danielsebas! Yes indeed, right this way: This error happens when your DADA2 is unable to join any of your reads. It is possible that your inclusion of so many low-quality nt positions is causing this. Any chance you can tighten up your trim/trunc params? This will depend on the length of the target region in question. Another option is to also proceed with only the forward reads.

Good luck!

[idupanloup (Isabelle Dupanloup)]

Hi Nicolas,
yes, I saw this explanation before.
but the quality scores of my reads are really good.
and this is strange to me that i need to trim my reads so much to be able to join them: denoise-paired with the following parameters: -p-trim-left-f 20 --p-trim-left-r 20 --p-trunc-len-f 100 p-trunc-len-r 150

don't you think so ?
best, Isa

[Nicholas_Bokulich]

@idupanloup

but the quality scores of my reads are really good.
and this is strange to me that i need to trim my reads so much to be able to join them: denoise-paired with the following parameters: -p-trim-left-f 20 --p-trim-left-r 20 --p-trunc-len-f 100 p-trunc-len-r 150

you must be truncating too much. Use longer truncation lengths to achieve sufficient joining between your paired-end reads. If this is not possible, you will need to process only your forward reads as if they are single-end reads.

[idupanloup (Isabelle Dupanloup)]

I tried to trim less bases : -p-trim-left-f 20 --p-trim-left-r 20 --p-trunc-len-f 80 p-trunc-len-r 100
but i get again the error message.
why should i have this type of problem if my reads are of good quality ?
i don’t get this.
thank you !

[Nicholas_Bokulich]

@idupanloup

why should i have this type of problem if my reads are of good quality ?

because they are not long enough for paired-ends to overlap after you trim them.

@idupanloup

I tried to trim less bases : -p-trim-left-f 20 --p-trim-left-r 20 --p-trunc-len-f 80 p-trunc-len-r 100

That's actually trimming more. The trunc-len defines the position at which truncation should occur, not how many bases to truncate off the 3' end.

[idupanloup (Isabelle Dupanloup)]

oops, sorry, yes, you’re right, thank you !