Importing single ended reads as 'SampleData[JoinedSequencesWithQuality]' does not work #2
Comments
|
Should be fixed in the 1.6.5 update. If not reopen this issue. |
|
Hello I am still getting a very similar error running itsxpress v1.7.2. Here is the command I am running after de-multiplexing. Do you think I would get a similar error with it not using the qiime2 plug in and using the tool itself? EDIT: I just read the open issues and I think that is where my issue is I have samples with no reads likely. Here is the error output: |
|
If you are using the newest versions of ITSxpress and Q2-itsxpress then this may be the same symptom but it likely has a different cause. From within your qiime conda environment run If you are up to date then the same error will be generated by any file that is empty after merging. This often happens when people include negative control files in their manifests. |
|
DOn't think the version is the problem I'm pretty sure the problem is that some of my samples didn't sequence well so after demultiplex they had less than 10 reads. I wrote a little work around script which exports out the demux.qza and creates a manifest of samples with more than 100 sequences then imports it back into qiime. This should fix them problem. |
|
@arivers I think I worked around the problem with files with no sequences by exporting the qiime artifact and reimporting only the files with more than 100 sequences; however, I've run into a different problem in which the output form ITSxpress is empty. I'm not entirely sure what is going on. When I take it through a otu vsearch pipeline I definitely have fungal sequences. Here is my pipeline: and this are the files I get out: You can see after the itsxpress step, the sequencing file drops significantly in size. The sequences were generated using the earth microbiome protocol. |
|
Any luck with this error? I am getting the same error as OP with latest qiime2 and ITSxpress 1.8.0. Like a previous poster, I limited the files with few reads by implementing a cutoff (simply by deleting samples from my manifest file) of samples with reads >100,200,500, 1000, and 5000. I am still getting the same error "File too small" |
|
It may be an issue with your sequences not merging well. can you run the qiime2 command in |
|
I am also thinking the same thing, but ITSxpress log file only gives the temporary fasta instead of the actual sample. it's probably just 1 samples throwing it off, but I'm not sure how to fix it. Either run BBmerge on individual samples, or possibly implement the newest version of vsearch that outputs an empty result file instead of an error. I actually opened up an issue here (USDA-ARS-GBRU/itsxpress#22) that includes the tail of my log file. I am also including the full debug output using the --verbose tag. |
|
I downloaded BBmerge and ran independently. I am running this analysis on some nasal swab samples that have a small bioload, so I initially implemented a very small read cutoff. However, even with the 500 read cutoff, I'm seeing several samples that have 100% ambiguity and no overlapping reads (i.e. see output below). `/Users/stricba1/bbmap//calcmem.sh: line 75: [: -v: unary operator expected [W::bgzf_read_block] EOF marker is absent. The input is probably truncated Pairs: 1530 Avg Insert: NaN Insert range: 999999999 - 0 [W::bgzf_read_block] EOF marker is absent. The input is probably truncated Pairs: 1619 Avg Insert: 245.3 Insert range: 122 - 482 I have run ITS analysis in the past where I have run through DADA2 and not through ITSxpress. However, these unpaired sequences will need to be omitted due to sequencing quality. I'm in the process of manually removing these reads with ambiguity near the pair, I will report back with results. |
|
I was able to run BBMerge outside of qiime2 and found that several of my samples did not pair. The new version of BBMerge seems to only send a warning, not an error when it runs into this issue. This way it will not break the pipeline. |
|
Hi @brittonstrickland if you are willing to share your data, I would like to see if I can replicate the problem and make the application more robust to this kind of error. Thanks, |
the plugin runs single-ended sequences that are of type 'SampleData[SequencesWithQuality]' fine but when I import the same sequences with the same manifest using the data type 'SampleData[JoinedSequencesWithQuality]'
qiime tools import --type 'SampleData[JoinedSequencesWithQuality]' --input-path /Users/rivers/Documents/itsxpress/tests/test_data/manifest2.txt --output-path demux2.qza --source-format SingleEndFastqManifestPhred33then run
trim-single --i-per-sample-sequences demux2.qza --p-region ITS2 --p-taxa F --o-trimmed trimmed2.qza --p-threads 2I get the following error:
The log error is
I exported the qza object and the fastq.gz file is not empty.
The text was updated successfully, but these errors were encountered: