combine two channels by sample string

[bio informatics]

Hi 

I have two channels and would like to combine them based on their sample names. However, keep failing. Here are the details:

Input Channels

ch_alignment

demux.Clontech_5p--bc1003_3p.flnc_clustered.sorted.sam
demux.Clontech_5p--bc1001_3p.flnc_clustered.sorted.sam
demux.Clontech_5p--bc1002_3p.flnc_clustered.sorted.sam


ch_clustered
demux.Clontech_5p--bc1001_3p.flnc_clustered.fasta
demux.Clontech_5p--bc1002_3p.flnc_clustered.fasta
demux.Clontech_5p--bc1003_3p.flnc_clustered.fasta

Desired output

[demux.Clontech_5p--bc1003_3p.flnc_clustered.sorted.sam, demux.Clontech_5p--bc1003_3p.flnc_clustered.fasta]
[demux.Clontech_5p--bc1001_3p.flnc_clustered.sorted.sam, demux.Clontech_5p--bc1001_3p.flnc_clustered.fasta]
[demux.Clontech_5p--bc1002_3p.flnc_clustered.sorted.sam, demux.Clontech_5p--bc1002_3p.flnc_clustered.fasta]

I tried "combine" and "filter" but it's not working for me

ch_alignment.flatten()

                        .combine(ch_clustered)

                        .filter {it[0].fileName.toString().split(".")[:-1].contains(it[1].fileName.toString().split(".")[:-2]) }

Any help is much appreciated. 

Cheers,
Mark

[Matteo Schiavinato]

I think you're not assuming that the two channels have the same order, therefore you must combine the two of them by some sort of matching key.

I think it won't be straightforward in this case, but you can tokenize the filenames in order to do so. I'd do it like this:

ch_alignment

.map{ it -> [ it.tokenize("."), it ] }

.map{ it -> [ [ it[0][0], it[0][1] ].join(".") , it[1] ] }

.set{ ch_alignment_token }

ch_clustered

.map{ it -> [ it.tokenize("."), it ] }
.map{ it -> [ [ it[0][0], it[0][1] ].join(".") , it[1] ] }
.set{ ch_clustered_token }

ch_alignment

.join(ch_clustered)

.map{ it -> it.flatten() }

.set{ ch_final }

output

ch_final.view()

[

[demux.Clontech_5p--bc1003_3p, demux.Clontech_5p--bc1003_3p.flnc_clustered.sorted.sam, demux.Clontech_5p--bc1003_3p.flnc_clustered.fasta]
[demux.Clontech_5p--bc1001_3p, demux.Clontech_5p--bc1001_3p.flnc_clustered.sorted.sam, demux.Clontech_5p--bc1001_3p.flnc_clustered.fasta]
[demux.Clontech_5p--bc1002_3p, demux.Clontech_5p--bc1002_3p.flnc_clustered.sorted.sam, demux.Clontech_5p--bc1002_3p.flnc_clustered.fasta]

]