Knuggets of Knowledge from the Stanford Bioinformatics for Microbiome Symposium

On Monday, 11th Sep 2017, I had the privilege of attending and being on a panel at the Stanford Bioinformatics for Microbiome Symposium. Ami Bhat and Ramesh Nair did an excellent job of bringing together a wide and eclectic group of presenters, including two undisputed leaders in their respective fields: Pavel Pevzner and Tom Slezak. Their deep knowledge and pedagogy was inspiring and refreshing. This was standing-room-only where someone stopped me to check if they could buy my badge!

Also wanted to give a shout out to my co-panelists: Elisabeth Bik, Duncan MacCannell, Mark Smith and Holly Ganz as well as the postgrads and grad students who presented posters.

Here are my key knuggets (the poop pun is purposeful) from the konference (in no particular order):

1. Deep NGS sequencing is not the answer to everything. PCR, microarrays and 16S are just fine sometimes
2. Abundance of taxa is inversely proportional to variation
3. Binary Trees are not the only available technique to speciate
4. Resolution matters for phenotypic determinants. This is where deep RNA-seq and other single-cell reactions and measurements matter.
5. Metagenome distribution is greater than 100 edges on De Bruijn Graph
6. The Rural Postman Problem is a subcategory of the Traveling Salesman Problem (useful for overlaps in sequencing)
7. The curse of dimensionality crops up often in microbiomics (among other ‘omices)
8. The human gut produces hundreds of cyclic peptides, most could be novel antibiotics
9. ~40% of self-reported subjects “not taking antibiotics for more than a year” show antibiotic molecules in stool… food sources? 5% to 10% of all supermarket food is fraudulent (source: The Atlantic Journal)
10. ~50% of self-reported IBS (irritable bowel syndrome) subjects have pseudomonas present
11. ICU patients have a very homogeneous microbiome, detrimental to recovery (Argonne UC study)
12. Reference Libraries are biased by commercial molecules
13. Bacteria use checkpoint inhibitors PD1
14. Rare bacteria are important to the phenotype
15. The microbiota may be the largest endocrine organ
16. CMOS rules. Computing on the edge will rule.

These events always give me a reality check of cutting me down to size, swallow more hubris and understand the value of humility… Comments welcomed!

## ?2017 Sanjay Joshi