Feature request: repLoad can read empty clone files that only have a header line.

[sciencepeak]

🚀 Feature

Let repLoad read the clone file that only has the header line.

Motivation

In an experiment of several samples, some samples might have no clones found by mixcr or trust4. Thereby, the clone files are empty except for a header line. I hope the empty files can also be read by the repLoad(), and the downstream analysis like overlap, clonality, and diversity, etc., can also be calculated by the immunarch functions, though there might be many zeros in the results. I think this function is important, because we want to the see clonotype change from one sample to another sample, even if the sample doesn't have have clones. That is the biological phenomena, and should be not skipped the immunarch.

Pitch

repLoad can read empty clone files that only have a header line.

Alternatives

Additional context

[vadimnazarov]

Hi @whitehilltea

Thank you so much for being so active, that helps a lot! This ticket is challenging to implement due to the internal immunarch architecture, to be completely honest with you. Can you provide more context for the case when your output files don't have any clones, please? What is your application for immune repertoires? Do you somehow additionally process FASTQ files with no clones to extract some other information? E.g., you use single-cell to look at transcriptomic profiles non-T/B cells.

[sciencepeak]

Hi, @vadimnazarov

I see the challenge.

The context is that we are studying the tumor infiltrating lymphocytes in the tumor tissues using bulk RNA-seq Fastq files. The upstream program is mixcr or trust4. The infiltration level of immune cells are different among samples at different stages under different treatments. We see clone abundance variances from very high to very low across different samples. so I think it is normal to find no clones in some samples.