Incorrect fastq paths in scRNA-seq sample sheet

Description of the bug

I am trying to use fetchngs to download scRNA-seq data from SRP348981 and prepare a sample sheet for input to the nf-core/scrnaseq pipeline. The scrnaseq pipeline requires columns sample, fastq_1, and fastq_2 in the sample sheet. The SRA records show three reads per spot: an index read (not required), read 1 (fastq_1), and read 2 (fastq_2). Read 2 is tagged as a biological read while the index and read 1 are tagged as technical reads. fetchngs with default options only downloads the fastq file for read 2. The path is added to the fastq_1 column of the sample sheet which is incorrect. Setting --download_method sratools results in successful download of all three fastq files but the sample sheet is still incorrect. The index read is added to the fastq_1 column and the fastq_2 column is still empty.

Is there a way to include both the read 1 and read 2 fastqs in the correct columns of the sample sheet with the current fetchngs version?

I have encountered this issue and others with several scRNA-seq datasets in SRA. Are there plans for future fetchngs releases to explicitly support --pipeline scrnaseq?

Command used and terminal output

nextflow run nf-core/fetchngs --input ./SRP348981_accession_list.csv --download_method sratools -profile docker

Relevant files

SRP348981_accession_list.csv
samplesheet.csv

System information

nf-core/fetchngs version 1.12.0

Start by reproducing the reported command with SRP348981_accession_list.csv and inspect the generated samplesheet.csv. Trace fetchngs metadata and FASTQ sample-sheet generation for the default and sratools download methods, then verify the result against the SRA records and the scrnaseq columns. Done means read 1 and read 2 are represented in the correct columns, with the index read handled appropriately.