Comment on Sally Pias – Molecular Modeling, Cholesterol, Oxygen Diffusion, (diabetes)

Dr Sally Pias | Molecular Modeling Changes How We Interpret EPR O2 Measurements | O2M Webinar Series

https://www.youtube.com/watch?v=QmcGZy_xZEA

Sally, thanks for the link to your paper. I found you on ResearchGate, but it only has the abstract (unless I missed something).

I like your visualizations and discussion. I see your paper on proteins and membranes, that helps clear up the question about proteins at the end. So I better understand your frustration.

I have not really looked for atlases or compilations of membranes in the human body. One term seems to be “lipidomic atlas” with such links as Lipidomic atlas of mammalian cell membranes reveals hierarchical variation induced by culture conditions, subcellular membranes, and cell lineages at https://pubs.rsc.org/en/content/articlelanding/2021/sm/d0sm00404a. They did not use absolute abundances but grouped by “phospholipid headgroups” and “total lipid unsaturation”.

You had to manually create your model. I tend to try to find the various environments of all membranes, then weigh each membrane effect on tissue and organ wide changes. I am looking at all 3D imaging methods on the Internet. I have followed ESR and EPR for the last 40 years or so. The modeling of proteins on membranes got a huge kick from covid, but those protein folding and membrane groups are a bit obscure and mostly only talk to themselves.

“membranes” (“EPR” OR “ESR”) “oximetry” “imaging” “diffusion” is giving 37,400 entry points, (Google, GMT 19 Dec 2021, 7:45 am)

So many groups and individuals have been struggling for a long time on each piece of this small puzzle. But other than talking to each other and writing papers that require a human reader to absorb and integrate, the diffusion of the results and applications is slow. About one thousandth or less of optimal.

I am writing this for my own notes on your talk. I watched these O2M videos and wonder why they are not getting wider connections.

Your video hardly got any views, but that might be because people working on EPR and ESR in the groups you are familiar with are not connected well.

“oximetry” “imaging” (“clinical” OR “in vivo”) has 642,000 entry points.

(“EPR” OR “ESR”) “molecular modeling” has 364,000 entry points

I use these kinds of searches as proxies for the size and impact of groups, then weigh them after closely investigating the actual groups, interconnections, and terms used. The equivalent would be something like correlating EEG, EMG, oximetry, MRI, ESRI, then using MRI spectroscopic imaging random samples to build a rough model of the whole.  (MRISI or ESRSI would be a way to rapidly characterize those lipidome groups.)

I think you might want to try to solve “cholesterol” “oxygen diffusion” “diabetes”. It only has 54,600 entry points. Just spend a few years to map and connect the groups – don’t try to do everything yourself. Find the groups doing “molecular modeling”, “machine learning” and show them where they can fit in.

And broaden your perspective to all (“in vivo” OR “clinical”) “3D imaging” with its 1.42 Million entry points. If all you do is find the groups, summarize where they fit into the overall interdiffusion of these two regions of the Internet, it will save decades of time by the current random diffusion.

I spent about 6 months and checked why “covid” diffusion is so slow. Partly it is unindexed and unlabeled and vastly duplicated materials on the Internet (now includes most of the published papers directly or indirectly). And moreso, the use of PDF and text formats that only human readers can access. If you ever played that children’s game of “telephone” or “whispers” where a child whispers into the ear of the next child in sequence until the last child blurts out what they heard and everyone laughs because it is always scrambled to incomprehensible – you know what is happening as PDF and text and other paper equivalents on the Internet are read by human eyeballs, matched to individual human memories and experiences, then manually transcribed into more papers.  It takes decades for even simple things to move one step, when the equivalent global methods take days or hours depending on the funding and focus of the groups involved.

Next time you talk, don’t be so nervous.

I wish I had more time. I am trying to organize all the 3D nerve activity imaging groups, for “spinal cord” “bypass” and “paralysis” “rehabilitation”.  It has thousands of major groupings, and hundreds of millions of connections. But not impossible. ESR has the penetration, but most groups are using really old technology for their “data engineering”.  Someone I know broke their neck, so I am taking a year to see if I can clean up that part of the Internet, while doing my regular Internet Foundation work on policies and methods.

“lipidomes” OR “lipidomics” OR “lipidome” has 3.11 Million entry points. When will these groups ever learn to self organize, index and curate?

Richard Collins, Director, The Internet Foundation