Catalog of viruses from human metagenomes reveals chronic disease associations | Hacker News Reader

Catalog of viruses from human metagenomes reveals chronic disease associations

pnas.org

146 pointsrch9 comments

With all the excitement around the study of the microbiome, its connection to the other organs (even brain), and role in a wide variety of diseases, I find it a bit daunting how little is understood about the other components of the microbiome beyond bacteria. The presence or absence of a specific phage or any other organism could be just as important if not more than relative balance of bacterial species.

I'm glad that these catalogs are growing and we are getting more datasets like this and the human protein atlas.

Remember when we thought the human genome project was going to "solve biology" :)

Agreed, especially available information around archaea and fungi

Prions are in a very different category then these 2, which are closer to humans phylogenetically then bacteria are. If they were present (some likely are) in these samples, they would be easily found in the NGS data.

And what about non-organisms like prions, which we barely grasp in terms of gray matter. What other types are there and what other havoc do they cause?

It should be stressed that association does not necessarily imply causation, and a variety of associative relationships between viruses and a given disease state are possible. For instance, virus abundance might simply be an epiphenomenon reflecting bacterial host abundance, the human genetics that predispose people to a disease might also provide a more favorable environment for the virus or its bacterial host, the external causes of a disease may create a more favorable environment for the virus, or the virus may contribute to the disease presentation in some way but ultimately does not cause the disease in isolation from other important factors. Verifying the associations we have detected with independent studies of the same diseases in additional populations will be key to understanding the extent to which the findings presented here are generalizable.

The author, Michael J. Tisza also shared some details in this twitter thread https://twitter.com/MikeTisza/status/1400528027285557248

If anyone is interested in other work on this topic, here is an article by one of my mentors, Mark Young:

https://pubmed.ncbi.nlm.nih.gov/27573828/

They found bacteriophages (viruses that infect bacteria) that were more common in people without IBS than with.

It explains the benefit of having a diverse microbiome because of increased adaptability of the bacteria to fighting off viruses

> Read data were downloaded from the National Center for Biotechnology Information’s (NCBI’s) Sequence Read Archive (SRA), including data from the Human Microbiome Project (34) and several other studies (25 Bioprojects in total) pursuing massively parallel sequencing of human metagenomic samples

Don't you need a statistical model accounting for possible sequencing batch effects, in that case? They may have been batched according to some scheme which correlates with the chronic diseases, and variations in the sequencing chemistry between batches could cause spurious associations.

Reply on news.ycombinator.com