[Cryptography] Cryptography of vertebrate virus defense

Arnold Reinhold agr at me.com
Mon Apr 27 21:24:40 EDT 2020


Back last century, I wrote a short paper analyzing one of the body’s defense mechanisms against viruses from a cryptographic standpoint. If you've ever wondered why humans reject organ transplants from other humans, something never encountered in nature, it’s because each of our cells has a mechanism that authenticates it to our immune system. The cells do this by sampling protein fragments that are created as the cell recycles old proteins, and presenting the selected fragments on the outside of the cell for inspection by immune cells. Only a small fraction of possible fragments are chosen, and the determination of which fragments is based on genes that are highly variable from individual to individual, effectively serving as a password. If a cell becomes infected by a virus, the virus protein fragments it presents can trigger an attack on it by immune system killer cells. (See https://en.wikipedia.org/wiki/MHC_class_I for background.)

However the immune system needs to be trained early in life to avoid recognizing normal cell proteins as foreign. The body has limited ability to perform this training and failures can result in auto immune diseases. My paper analyzed this trade off between fragment selection fraction, effectively password entropy, and the risk of training failures. Here is a link for anyone interested:

    https://www.researchgate.net/publication/239937720_On_the_Function_of_MHC-Antigen_Specificity

Arnold Reinhold


More information about the cryptography mailing list