Adapting nf-core/sarek for Ion Torrent Single-End Data Analysis

Question

[Nour-eddine]

Hello !

I'm attempting to utilize the Sarek pipeline for analyzing Ion Torrent data, but I'm facing challenges with configuring single-end files. Unlike Illumina, the Ion Torrent data I have comes in the form of a single FASTQ file per sample.

My issue lies in adapting the sample sheet to accommodate these single-end data. According to the documentation, the sample sheet structure is designed for paired-end data, whereas in my case, I only have one FASTQ file per sample.

Here's an example of my data structure:

patient,sample,lane,fastq_1

patient1,test_sample,lane_1,test.fastq.gz

I've created this simplified configuration for my single FASTQ file, but I'm wondering if there's a specific method to correctly adapt the sample sheet for Ion Torrent single-end data.

Additionally, are there specific parameters or configurations within the Sarek pipeline that need adjustment to support this single-end data type from Ion Torrent?

Any help or guidance in correctly adapting the nf-core/sarek configuration for Ion Torrent single-end data would be greatly appreciated!

Thank you very much for your assistance!

Best regards,

Nour-Eddine

Answer 1

[Dave Carlson]

Sarek can only be run with paired end data, so you will not be able to use it with your single-end Ion Torrent data. In the past, when people have asked about using single end data with Sarek, the developers have suggested that the Eager pipeline might be a possible alternative.

[Nour-eddine]

Thank you for your clear and helpful response regarding Sarek and single-end data. I appreciate your help!

[Nour-eddine]

I'm considering converting my Ion Torrent data into R1 and R2 files to suit Sarek. Do you think this conversion would be suitable in this context, or are there other considerations I should keep in mind?

[Dave Carlson]

I would definitely not recommend trying to convert single end fastq files into pseudo paired end data. In my opinion it would be better to use/adapt a different pipeline or simply write your own.

Answer 2

[Friederike.hanssen]

Hi! Yes as @Dave Carlson mentioned, nf-core/eager might be a suitable alternative. In addition, several people have been testing out sarek with single end data from the duplicate marking step. If this works well, we could consider adding more aligners.

here is a link to the discussion: https://nfcore.slack.com/archives/CGFUX04HZ/p1701406781438139