PREreview of Long-read sequencing of Mycobacterial tuberculosis is comparable to short-read sequencing for antimicrobial resistance prediction and epidemiological studies

Summary

This preprint compares long-read ONT sequencing with short-read Illumina sequencing for Mycobacterium tuberculosis. The authors find very high agreement between the two platforms for drug resistance prediction, lineage identification, and SNP-based transmission clustering. I think this is an important and timely study because it suggests ONT may now be accurate enough to use alongside Illumina in clinical and public health settings.

Main contribution

The main contribution is that ONT and Illumina data may be similar enough to combine for large-scale TB analyses, as long as sequencing quality is high. This is valuable because ONT could allow faster and more portable sequencing in high-burden settings, which could help with outbreak tracking and treatment decisions.

Major comments

One limitation is that Illumina is treated as the reference standard. This is understandable, but it means disagreements are mostly interpreted as ONT errors. The authors should acknowledge that some discordant calls could reflect variants missed by Illumina or consider validating a subset of disagreements.

Another limitation is that only high-quality sequencing samples were included. This makes the comparison clearer, but it may overestimate how well ONT would perform in routine settings where sample quality varies. The authors could discuss how lower-quality samples might affect resistance prediction and transmission clustering.

The mixed infection results also need more explanation. Some lineage differences came from mixed lineages being detected by one platform but not the other. Since mixed infections may occur in real-world TB surveillance, the authors should clarify how each platform handles them and what this means for interpreting results.

Finally, the samples come from South Africa and Vietnam only. These are important TB settings, but the authors should discuss whether the findings are likely to generalize to other regions and lineages.

Minor comments

The variant frequency cutoffs, such as the 5% threshold, could use clearer justification.

Some figures could be easier to follow with clearer labels or more detailed legends.

One of the biggest reasons I find this study exciting is that ONT is far more portable than traditional sequencing platforms, so if it can produce results close to Illumina, it could make high-quality TB sequencing much easier to use in the places where it is needed most.

Overall assessment

Overall, I think this is a strong and useful preprint. It makes a convincing case that ONT sequencing can perform similarly to Illumina for important TB applications. I like this preprint a lot and think it has real public health value, especially because ONT could support faster and more portable sequencing in high-burden settings. Addressing the limitations above would make the conclusions stronger and more useful for real-world implementation.

Competing interests

The author declares that they have no competing interests.

Use of Artificial Intelligence (AI)

The author declares that they did not use generative AI to come up with new ideas for their review.