How to work with multiple time point/samples for a patient? read CSV

Ask for help
1

Hi there,

I work with patient data (whole exome, RNA), where there can be multiple samples from different time points.

Patient_ID Sample name DNA_N_R1 DNA_N_R2 DNA_T_R1 DNA_T_R2 RNA_T_R1 RNA_T_R2
patient1 patient1-3 DNA1_N_R1 DNA1_N_R2 DNA_T_T01_R1 DNA_T_T01_R2 RNA_T_T01_R1 RNA_T_T01_R2
patient1 patient1-4 DNA1_N_R1 DNA1_N_R2 DNA_T_T02_R1 DNA_T_T02_R2 RNA_T_T02_R1 RNA_T_T02_R2
patient1 patient1-5 DNA1_N_R1 DNA1_N_R2 DNA_T_T03_R1 DNA_T_T03_R2 RNA_T_T03_R1 RNA_T_T03_R3

the normal in DNA will remain constant but tumor will vary.
There’s no RNA normal, only tumor.

How do I create any channel/data structure to manage this type?

Earlier I was working with one row per patient and was under the impression unique tumor, normal DNA samples and only one timepoint for RNA for which I had following code:

Channel.fromPath( csv_file )
.splitCsv( )
.multiMap { row →
def rna_reads=tuple(file(row[5]), file(row[6]))
def tumor_reads = tuple( file( row[3]) ,file(row[4]))
def normal_reads = tuple(file(row[1]),file(row[2]) )

        tumor:
             tuple( row[0], tumor_reads )
        normal:
             tuple( row[0], normal_reads )
            rna:
            tuple(row[0],rna_reads)
    }.set{samples}

Then they can be sent for analysis:

    align(both.out.reads_tumor)
    ariba(align.out.aligned_star)
    markduplicates(align.out.aligned_star)
    featurecounts(markduplicates.out.markduplicate_bam)

This worked fine with following input data:

Patient_ID DNA_N_R1 DNA_N_R2 DNA_T_R1 DNA_T_R2 RNA_T_R1 RNA_T_R2
sample1 DNA1_N_R1 DNA1_N_R2 DNA1_T_T01_R1 DNA1_T_T01_R2 RNA1_T_T01_R1 RNA1_T_T01_R2
sample2 DNA2_N_R1 DNA2_N_R2 DNA2_T_T02_R1 DNA2_T_T02_R2 RNA2_T_T02_R1 RNA2_T_T02_R2
sample3 DNA3_N_R1 DNA3_N_R2 DNA3_T_T03_R1 DNA3_T_T03_R2 RNA3_T_T03_R1 RNA3_T_T03_R3