Inquiry About Virus Genome Detection in Human RNA-seq Data

[zhuhonglin2022]

Dear Authors,

Thank you for your impressive and wonderful work.

I was wondering if your model can detect viral genomes in human RNA-seq data. We have tested several tools but haven't found one that performs very well. If your model can achieve this, it would be highly beneficial for biologists.

Looking forward to your insights.

Best regards,

[garykbrixi]

Hi @zhuhonglin2022
Thanks for your question and interest

Evo 2 was not trained on viral RNA-seq data but was trained on DNA from eukaryotes including humans. It may be possible to do viral genome detection in RNA-seq based on this, as it has high perplexity on sequences from eukaryotic viruses

[xzx0554]

Hi, I have fine-tuned an Evo2 (just 500M) on a virus dataset. If you are interested in that, you could contact me.

[zhuhonglin2022]

That's great!
We have bulk RNA-seq and single-cell RNA-seq data—would it be possible to detect viral genomes using your model?
If so, would you be open to collaborating on this?

Best wishes,
Honglin

[xzx0554]

[email protected]