Does ultraplex handle barcodes that might be reversecomplement of each other

[Rohit-Satyam]

Hi developers

I just realised that some of the barcodes in my samples are the reverse complement of some other barcodes. Is Ultraplex capable of handling such scenarios? I am worried about these samples because I suspect they might get mixed reads!!

ultraplex -i TW_1.clean_1.fastq.gz -i2 TW_2.clean_2.fastq.gz -d 1mismatchallowed_demux/ -t 32 -m5 1 -q 0 -b barcodes.csv -u --keep_barcode

sample_22	TCCTGAGC	<->	sample_14	GCTCAGGA
sample_13	CTATCGCT	<->	sample_1	AGCGATAG
sample_6	ATCACGAC	<->	sample_18	GTCGTGAT
sample_15	GTACTGAC	<->	sample_17	GTCAGTAC
sample_2	AATGAGCG	<->	sample_11	CGCTCATT
sample_10	CATGCCTA	<->	sample_20	TAGGCATG

[Rohit-Satyam]

I also looked at the first 12 bases first 1M of reads and I realised that most of the barcodes appears in the reads from R2 file. However, R1 reads have small proportion of barcodes too I have reached out to the person who generated the libraries to understand how the libraries were prepared. Does ultraplex look for barcodes in R1 reads? Or R2 reads or either or both?

[Delayed-Gitification]

The best explanation of the way ultraplex works is our publication in Wellcome Open Research, which has some diagrams too