This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Whole Genome Sequencing IG regions

I am looking at some sequencing data in https://doi.org/10.1038/s41588-024-01853-0 and am particularly interested in the loci around the immunoglobulin regions, on Chromosomes 2,14 & 22. They use Illumina HiSeq2000 & HiSeq2500 and a large chunk of somatic mutations found are in or around the immunoglobulin genes, and many not in V genes. Worth noting this is from plasma cells, that will have undergone recombination and somatic hypermutation as these loci. Is this data interesting or did more need to be done to ensure correct mapping of reads etc?

Thanks in advance

immunoglobulin wgs

I apologise for the poor wording. My worry is that in a region as heavily re-arranged and mutated as this, that the reference genome may become quite a lot less useful when carrying out alignment and variant calling. Given a read may contain information that has been recombined, that would present a challenge aligning for example. There is a piece of software called IgCaller that specialises in this (https://doi.org/10.1038/s41467-020-17095-7). However, since I don't have the raw data I cannot redo any of the work. So perhaps instead of the word 'interesting' I should have said, are these calls reliable in the context of lymphoid cells?

Apologies for the poor wording and thanks for taking the time to comment. By 'interesting', I really mean 'reliable'. Given the nature of the region with recombinations and SHM in a B-cell, I would have thought that alignment and variant calling methods may be somewhat hampered. For example, if a read contains information that has been recombined then I would have thought mapping to the reference would be impossible. Unfortunately I don't have the raw data so cannot redo anything but there is a piece of specialist software called IgCaller for this.

The question is what you are after. Is this a disease setup, or...well, what is this all about. Generally, these regions are hotspots of hypermutation, so a lot of varians are expected. Is there a scientific question at hand or are you fishing for a project?

Sorry for the poor wording of the original post, also should say I really appreciate people taking a moment to look at my post. My main concern is that given there are SHM & recombination in these regions, are the calls I am seeing reliable or should I take them with something of a pinch of salt?

are the calls I am seeing reliable or should I take them with something of a pinch of salt?

Not sure what calls you are referring to but the sequence basecalls should be reliable considering this appears to be a recent paper (though HiSeq 2K and 2500 are older sequencers and the data could be several years old before it got published in 2024, we have no way to know).

You will need to decide what kind of error you are willing to allow when you align the reads (this is not always an easy setting with aligners and may not be always available) and if the end result suits your purpose.

1 answer

The sequencing data from the study (DOI: 10.1038/s41588-024-01853-0) is interesting because it reveals a high frequency of somatic mutations in and around immunoglobulin genes on chromosomes 2, 14, and 22 in plasma cells from multiple myeloma patients. These mutations align with expected biological processes, as plasma cells undergo VDJ recombination and somatic hypermutation, which generate diversity in immunoglobulin loci. The fact that many mutations occur outside V genes is noteworthy, as somatic hypermutation typically concentrates in variable regions but can extend to adjacent areas, potentially reflecting off-target effects or additional mutational mechanisms in malignant plasma cells.

However, the study employed standard read mapping with BWA to the GRCh37 reference genome and variant calling via tools like Seurat, Strelka, and MuTect, without documented specialized handling for immunoglobulin regions. This approach may introduce mapping errors, as the reference genome represents germline configurations, not the rearranged loci in individual plasma cells. Misalignments can result in false positive variant calls or underestimation of true hypermutations.

To ensure correct mapping, additional steps are required, such as using immunoglobulin-specific alignment tools like IgBLAST or MiXCR, or constructing personalized references incorporating VDJ segments. Validation through orthogonal methods, like targeted amplicon sequencing, would also strengthen reliability. Without these, interpret the immunoglobulin-associated mutations cautiously, as they may include artifacts alongside genuine events.

Regarding the comments on this post, the base calls from Illumina HiSeq2000 and HiSeq2500 are generally reliable, but alignment tolerances must be adjusted for hypervariable regions to minimize errors. The scientific context is multiple myeloma genomics, where immunoglobulin mutations are passenger events rather than drivers, yet they provide insights into disease biology.

Kevin

Log in to answer this question.