Smythies John
Laboratory for Integrative Neuroscience, Center for Brain and Cognition, University of California, San Diego La Jolla, CA, USA.
Front Mol Neurosci. 2015 Jul 23;8:38. doi: 10.3389/fnmol.2015.00038. eCollection 2015.
Until recently it was held that the neurocomputations conducted by the brain involved only whole neurons as the operating units. This may however represent only a part of the mechanism. This theoretical and academic position article reviews the considerable evidence that allosteric interactions between proteins (as extensively described by Fuxe et al., 2014), and in particular protein vibrations in neurons, form small scale codes that are involved as parts of the complex information processing systems of the brain. The argument is then developed to suggest that the protein allosteric and vibration codes (that operate at the molecular level) are nested within a medium scale coding system whose computational units are organelles (such as microtubules). This medium scale code is nested in turn inside a large scale coding system, whose computational units are individual neurons. The hypothesis suggests that these three levels interact vertically in both directions thus materially increasing the computational capacity of the brain. The whole hierarchy is thus similar to three nested Russian dolls. This theoretical development may be of use in the design of experiments to test it.
直到最近,人们一直认为大脑进行的神经计算仅涉及整个神经元作为操作单元。然而,这可能仅代表了机制的一部分。这篇理论性和学术性的立场文章回顾了大量证据,即蛋白质之间的变构相互作用(如富克斯等人在2014年广泛描述的那样),特别是神经元中的蛋白质振动,形成了作为大脑复杂信息处理系统一部分的小规模编码。接着展开论证,提出蛋白质变构和振动编码(在分子水平上运作)嵌套在一个中等规模的编码系统中,其计算单元是细胞器(如微管)。这个中等规模的编码又嵌套在一个大规模编码系统中,其计算单元是单个神经元。该假说表明,这三个层次在两个方向上垂直相互作用,从而实质性地提高了大脑的计算能力。因此,整个层次结构类似于三个嵌套的俄罗斯套娃。这一理论进展可能有助于设计实验来对其进行检验。