Information geometry and divergences

https://franknielsen.github.io/IG/

David Sinclair's information theory of aging

lifespan - Why we age and why don't have to. David Sinclair 

Sinclair claimed that he had the idea of the Information Theory of Aging on October 28, 1996, page 35, with a hand-written title provided therein. 

Sinclair presented the arguments that DNA mutation is not a cause of aging, but the chaos air of epigenome such as DNA methylation is the major cause of aging. 

Sinclair claims that "aging is a loss of information". 

Hong think this is NOT right, at least based on Shannon's information formula,  H = - sum( p* log(p)) , because in chaotic old aging, there would be more states and more randomness, so information entropy would increase not decrease. So, it seems that Shanon's H would increase during aging. 

blogs on Sinclair's lifespan book: 

https://hplus.club/blog/a-summary-of-david-sinclairs-information-theory-of-aging/

https://medium.com/intuitionmachine/deep-learning-and-solving-aging-21eaaa6eec8b#:~:text=David%20Sinclair's%20%E2%80%9Cinformation%20theory%20of,an%20organism%20eventually%20stops%20growing.

A good contradictory riddle in Sinclair's information theory of aging and reprogramming for youth:  

https://hplus.club/blog/david-sinclairs-level-3-reversing-aging-with-the-yamanaka-factors/

Al-hasmi talk, conformational penalty

Conformational penalties: The other half of molecular recognition

At the most fundamental level, living organisms are the product of biomolecules interacting with one another through a process commonly referred to as molecular recognition.  To understand how biomolecule interact with one another, we need a framework that describes those properties of the biomolecules that determine their binding affinities and specificities.  Our current understanding dates back six decades ago when Linus Pauling proposed that specificity is achieved through the structural complementarity of the binding partners.  This concept has been reinforced over the decades thanks to advances in the determination of high-resolution structures of biomolecules by X-ray crystallography and cryoEM.  Static structures only carry information regarding one half of the molecular recognition equation, which I will refer to as income.  This half describes the favorable contacts formed upon complex formation.  The second half, which has received much less attention, I will refer to as income tax.  It represents the energetic cost associated with changing the structure of a biomolecule from one form to another when binding a partner molecule.  Unlike income, the income tax half of the molecular recognition equation can only be determined experimentally through an ensemble description of biomolecules as a probability distribution of many different conformations.  I will argue that income tax, and mechanisms for tax evasion, are ubiquitous in biology and disease, drawing on DNA replication as a primary example. 

https://www.nature.com/articles/s41586-020-2843-2

https://sites.duke.edu/alhashimilab/research/

free energy tax

distribution of the sub-states really affect taxation of free energy. Mentioned to use MD to run and get distribution of molecules, keep the caveat that it might not reflect experimental results. 

TF factors and motifs:: specificity and affinity are generally correlated, but can be altered. 

free engery and probability landscape is connected by logrithm

Tax impairs DNA replication forks and increases DNA breaks in specific oncogenic genome regions

Hassiba Chaib-Mezrag, Delphine Lemaçon, Hélène Fontaine, Marcia Bellon, Xue Tao Bai, Marjorie Drac, Arnaud Coquelle, and Christophe Nicot. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4168069/

"Free energy minimisation is equivalent to maximising the mutual information between sensory states and internal states that parameterise the variational density (for a fixed entropy variational density).^{[11][better source needed]} This relates free energy minimization to the principle of minimum redundancy^[25] and related treatments using information theory to describe optimal behaviour." 

This is related to our cross-entropy work on aging analysis 

Kullback-Leibler divergence, or relative entropy

https://en.wikipedia.org/wiki/Relative_entropy

active inference by Karl Friston

Re Friston: a good intro talk: https://www.santafe.edu/events/me-and-my-markov-blanket

what’s the authors name of the self replicating machine?

von Neumann

 The book that was cited is called "The theory of self-reproducing automata"

Here’s Art Burk’s review of von Neumann: http://walterfontana.zone/wp-content/uploads/2020/12/Burks-1969.pdf

https://www.sciencedirect.com/science/article/pii/S2405471218300577

 This paper does exactly that on the Lac repressor

toread, Effects of reciprocity on random walks in weighted networks Zhang, Li, Scientific Report

Effects of reciprocity on random walks in weighted networks
Zhang, Li, Scientific Reports