Skip to main content

Microbiomes by Tal Korem

Lecture was about the microbiome and its importance and association with multiple chronic and complex diseases (I love this topic!) Larger genome in microbiome than in us. What's really exciting is that you can actually modify your microbiome. Importance + Modifiability = Biomedical Promise.

Overview of microbiome in clinical settings​

  • You can start with how drugs interact with the microbiome (usual guesses and don't forget that some drugs are _anti_microbial; this doesn't necessarily mean that they're deletrious).
  • Another avenue is immunotherapy. It has been demonstrated that microbiomes have mediated responses to immunotherapy.
  • Another use as diagnostic tools (e.g. colorectal cancer and early preterm birth; may be better than blood or other draws!)
  • Modulation, Personal Medicine (e.g. personalized diets outperform standard of care).
  • Postbiotics: can you use MB as a discovery playground for future therapies? E.g. ALS mice have a distinct microbiome community.

Basics of Analysis: Data, Information, Analysis​

Data​

16s-rRNA gene sequencing. Why this rRNA gene? It's an extremely conserved as a phylogenetic marker. You can use this to really understand bacteria. The person who discovered this actually created the first Tree of Life and discovered archaea. It's not present in human cells, but in all prokaryotes, doesn't like being shared between organisms, and is 'large enough' to have information. But really, it's because you see areas of high and low entropy (genetic info) evenly distributed which is awesome for PCR! You can design good primers because it's conserved enough and it's variable enough to indicate phylogeny. Discussion on caveats; no free lunches here. Stool Sample -> DNA -> PCR -> 16S Amplicons -> Next Gen Seq -> Data. This has higher sensitivity.

Another way is Metagenomics and de novo assembly. Why sequence all genes? Sequencing became cheaper; No need for PCR. You then need to use a database of reference genomes for mappings. This has higher specificity.

There are other readouts, other "omics" you can get out of microbiome communities.

Now you have to be careful about what you're sampling and where it came from. Half-life of RNA is minutes. Think of transporting stool samples. DNA reflects things more proximal, RNA more distal (your rectum).

Information​

Taxonomic relative abundances (types of bacteria). Functional relative abundances (types of genes).

Problem with relative abundances is that they are compositional (sum to one). Proportions will stay static and won't tell you what happened to a starting community after some effect/intervention/experiment. Another is that there is a strong measurement bias in microbiome experiments: you can send the same samples to the same lab six months apart and get wildly different results. Standardization is a problem. Contamination is a problem.

Bacterial growth dynamics cannot be inferred from relative abundances! A/B are 50%/50%, 12h later, 30%/70%. What happened?

  • A died, B thrived
  • A stayed same, B thrived
  • A and B grew but rates were different

Analysis​

Core question: how can you use associations to inform interventions? So if you have a study that specifies that, for condition X, you have a 'bad' microbiome and a 'good' microbiome, what would you do as a clinician? Fecal transplants! They're used pretty regularly and for "C. Diff" infections. You can do several others but the efficacy is related to how stable the microbiome is. In T2 Diabetes, you won't get much bang for your buck with transplantation because even with the chronic condition, the microbiome is complex, stable, and doing its job!

OK, another way (literally) is probiotics. There are a lot of conflicting reports on these, lots of studies dispute their efficacy. Little muscosal presence and no long-term colonization (you poop them out mostly). You can make things worse too! Your microbiome might need to actually recover from the presence of a foreign microbiome (especially if it's stable as described earlier!)

Fecal Transplants: Capsule with feces up the derriere or through the nose.