Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Toward Biochemical Probabilistic Computation

Abstract : 1 Abstract To account for the ability of living organisms to reason with uncertain and incomplete information , it has been recently proposed that the brain is a probabilistic inference machine , evaluating subjective probabilistic models over cognitively relevant variables. A number of such Bayesian models have been shown to account efficiently for perceptive and behavioral tasks. However , little is known about the way these subjective probabilities are represented and processed in the brain. Several theoretical proposals have been made , from large populations of neurons to specialized cortical microcircuits or individual neurons as potential substrates for such subjective probabilistic inferences. In contrast , we propose in this paper that at a subcellular level , biochemical cascades of cell signaling can perform the necessary probabilistic computations. Specifically , we propose that macromolecular assemblies (receptors , ionic channels , and allosteric enzymes) coupled through several diffusible messengers (G-proteins , cytoplasmic calcium , cyclic nucleotides and other second messengers , membrane potentials , and neurotransmitters) are the biochemical substrates for subjective probability evaluation and updating. On one hand , the messengers ' concentrations play the role of parameters encoding probability distributions ; on the other hand , allosteric conformational changes compute the probabilistic inferences. The method used to support this thesis is to prove that both subjective cognitive probabilistic models and the descriptive coupled Markov chains used to model these biochemical cascades are performing equivalent computations. On one hand , we demonstrate that Bayesian inference on subjective models is equivalent to the computation of some rational function with nonnegative coefficient (RFNC) , and , on the other hand , that biochemical cascades may also be seen as computing RFNCs. This suggests that the ability to perform probabilistic reasoning is a very fundamental characteristic of biological systems , from unicellular organisms to the most complex brains. Living organisms survive and multiply even though they have uncertain and incomplete information about their environment and imperfect models to predict the consequences of their actions. Bayesian models have been proposed to face this challenge. Indeed , Bayesian inference is a way to do optimal reasoning when only uncertain and incomplete information is available. Various perceptive , sensory-motor , and cognitive functions have been successfully modeled this way. However , the biological mechanisms allowing animals and humans to represent and to compute probability distributions are not known. It has been proposed that neurons and assemblies of neurons could be the appropriate scale to search for clues to probabilistic reasoning. In contrast , in this paper , we propose that interacting populations of macromolecules and diffusible messengers can perform probabilistic computation. This suggests that probabilistic reasoning , based on cellular signaling pathways , is a fundamental skill of living organisms available to the simplest unicellular organisms as well as the most complex brains .
Complete list of metadata

Cited literature [42 references]  Display  Hide  Download
Contributor : Pierre Bessiere Connect in order to contact the contributor
Submitted on : Monday, November 9, 2015 - 11:38:57 AM
Last modification on : Wednesday, May 19, 2021 - 11:58:08 AM
Long-term archiving on: : Wednesday, February 10, 2016 - 10:17:37 AM


Files produced by the author(s)


  • HAL Id : hal-01226297, version 1


Jacques Droulez, David Colliaux, Audrey Houillon, Pierre Bessière. Toward Biochemical Probabilistic Computation. 2014. ⟨hal-01226297⟩



Record views


Files downloads