Demultiplexing dual barcodes with older data

[raw937]

Within one fastq I have 16 potential barcodes

For example: In bold dual barcode

#R1 read

@SOLEXA1_0069_FC:3:1:1673:948#ACAGTG/1
GACT AACCGGATTAGATACCCTGGTAGTCCACGCCGTAAACGATGAATGTTAGCCGTCGGGCAGTATACTGTTCGG
+
BMMQNTWSWWb_____b_bb__________Y_________YYYYY[[[Y[__________XXRWXVVVVTYYYYYT

#R2 read

@SOLEXA1_0069_FC:3:1:1673:948#ACAGTG/2
CTGA AGGGTTGCGCTCGTTGCGGGACTTAACCCAACATCTCACGACACGAGCTGACGACAGCCATGCAGCACCTGT
+
ghgaggfghhhhhhhhhhghhhhhhhhhhfhhhghfWffch[hhgahhedffddR[^W^Zc^\_cac[Wb]^W^

The barcodes are here (16 possible):

R1*R2

TCAGTCAG
CTGACTGA
TCAGGACT
GACTGACT
AGTCAGTC
GACTTCAG
GACTAGTC
GACTCTGA
TCAGCTGA
AGTCTCAG
AGTCGACT
CTGAAGTC
CTGAGACT
AGTCCTGA
TCAGAGTC
CTGATCAG

need some help asap.

[ebolyen (Evan Bolyen)]

Hey @raw937,

I hate to say this, but QIIME 2 still doesn’t understand dual-indexed (combinatoric) barcodes >_<
It’s been on our list for basically forever, but it just hasn’t happened yet.

In the meanwhile, if there’s a tool anyone is aware of for demultiplixing this, you could use that and then import the demultiplexed reads in QIIME 2 without any issues.

Does anyone have a good way of handling dual-indexed reads?

[colinbrislawn (Colin J Brislawn)]

Gentlemen,

Looks like the demultx from the ea-utils package might be able to do this…

I’ve never worked with dual-indexed reads before, but it might be worth a shot.

Colin