Custom Adapter Sequence in Kneaddata

Infrastructure and utilities KneadData
1

Hello, I would like to use kneaddata with a custom adapter sequence, rather than the ones pre-installed (ie. TruSeq2, TruSeq3, NexteraPE). I have the sequence set up as a .fa file. But I can’t figure out how to get kneaddata to use my adapter FASTA file.

Additionally, I tried to use trimmomatic to trim the adapters then pass the sequence off to kneaddata with the flag --bypass-trim since I already trimmed the adapters:

kneaddata --input1 /path/to/trimmed/sequence/test_001_S77_L004__trimmed_1P.fastq.gz \
    --input2 /path/to/trimmed/sequence/test_001_S77_L004__trimmed_2P.fastq.gz \
    -db /path/to/reference_db/mouse \
    --run-trim-repetitive \
    --run-fastqc-start \
    --run-fastqc-end \
    --bypass-trim \
    --threads 32 \
    --output /writeable/path/kneaddata/

but I ended up with an error: ERROR: Unable to write file: /writeable/path/kneaddata/reformatted_identifiers4yt3gtxj_decompressed_ijjeqjgd_test_001_S77_L004__trimmed_1P

Any help is appreciated!

I am using
[kneaddata v0.12.2]

2

Update: So I figured out what was causing the error. Turns out the Kneaddata requires the input files to have the format of filename.R1.fastq.gz and filename.R2.fastq.gz despite using the --input1 and --input2 flags. Hopefully if anyone comes across this issue, this helps!

But I still am not sure how to specify custom adapter sequences in kneaddata or if that’s even possible.

3

There are a few non-default adapter options provided out of the box: GitHub - biobakery/kneaddata: Quality control tool on metagenomic and metatranscriptomic sequencing data, especially data from microbiome experiments.
If your sequences are not covered by any of these, you should be able to manually set the trimmomatic options with the “–trimmomatic-options” flag. Note that using this flag removes the default options, so you’ll need to specify the default options “SLIDINGWINDOW:4:20 MINLEN:50” in addition to your adapter specific options if desired.