split_libraries_fastq.py with demultiplexed samples warning: "Mean of empty slice.", RuntimeWarning

[Ricardo Ramiro]

Dear all,

I was running the split_libraries_fastq.py on demultiplexed samples, with the following command:

split_libraries_fastq.py -i P3_L001_R1_001.fastq,P2_L001_R1_001.fastq,P1_L001_R1_001.fastq,O3_L001_R1_001.fastq,O2_L001_R1_001.fastq,O1_L001_R1_001.fastq,N3_L001_R1_001.fastq,M3_L001_R1_001.fastq,M2_L001_R1_001.fastq,M1_L001_R1_001.fastq,J2_L001_R1_001.fastq,H3_L001_R1_001.fastq,H1_L001_R1_001.fastq,G3_L001_R1_001.fastq,G1_L001_R1_001.fastq,F2_L001_R1_001.fastq,F1_L001_R1_001.fastq,E3_L001_R1_001.fastq,E2_L001_R1_001.fastq,Ctr_L001_R1_001.fastq,C3_L001_R1_001.fastq,C2_L001_R1_001.fastq,C1_L001_R1_001.fastq,B3_L001_R1_001.fastq,B2_L001_R1_001.fastq,B1_L001_R1_001.fastq,A3_L001_R1_001.fastq,A2_L001_R1_001.fastq,A1_L001_R1_001.fastq --sample_ids P3,P2,P1,O3,O2,O1,N3,M3,M2,M1,J2,H3,H1,G3,G1,F2,F1,E3,E2,Ctr,C3,C2,C1,B3,B2,B1,A3,A2,A1 -o slout_not_multiplexed_q20/ -q 19 --barcode_type 'not-barcoded'

However, at the end I got the following warning:

/usr/local/Envs/qiime/local/lib/python2.7/site-packages/numpy/core/_methods.py:59: RuntimeWarning: Mean of empty slice.
 

I have looked through the posts in the forum, but most of the answers seem to relate this warning to barcodes. However, my sample is not barcoded.

So my questions are:

- is this warning a problem? (the seqs.fna file has been produced)

- if it is, does anyone have a solution? (i have attached the log file).

Thanks for any help,

Ramiro

[TonyWalters]

Hello Ramiro,

It looks like you got that warning because the last input file, A1_L001_R1_001.fastq had no reads written (also it only had 2 reads in it, which is really low). It looks like you had a few other input fastq files that were really low sequence count, but this was the only zero output one. It's probably fine to ignore the warning in this case, since you had other input fastq files that wrote numerous sequences. It may be worth investigating if other preprocessing (I don't know if you had stitched these reads?) and make sure there weren't issues with the barcodes specified when sequencing (may account for low reads, or may just be a few failed samples).

Input file paths
Sequence read filepath: A1_L001_R1_001.fastq (md5: 2843a404a652ebd3f188107b6fd2f379)
Quality filter results
Total number of input sequences: 2
Barcode not in mapping file: 0
Read too short after quality truncation: 2
Count of N characters exceeds limit: 0
Illumina quality digit = 0: 0
Barcode errors exceed max: 0

Result summary (after quality filtering)
Median sequence length: nan
A1	0

Total number seqs written	0