218

effects (most of the time), but can also lead to negative effects (rarer) or to positive effects

(even rarer) – and rarely detected, because it then does not lead to any complaints, on the

contrary. For example, we investigate an SNP that has an effect on the psyche. A sequence

analysis first provides information on which DNA segments with a known function the

gene resembles. This is followed by further analyses to determine whether the SNP also

appears as RNA in the brain or even as a protein. Based on this, the RNA or protein struc­

ture is then analysed and how the difference of one nucleotide affects this. Subsequently,

one can try to predict (from databases and with prediction algorithms, a first start is the

tool STRING at the EMBL: https://string-­db.org) with which proteins there are interac­

tions here. In this way, one can determine step by step what effects this small change has.

Of course, there is also everything in between. Smaller or longer insertions or deletions

in the genome sequence on the affected chromosome and also very large modifications,

such as additional chromosomes (the best known is trisomy 21, Down syndrome) or incor­

rectly assembled chromosomes (“translocations”). The database “Online Mendelian

Inheritance in Man” (OMIM; https://www.omim.org) provides a detailed overview. Genes,

proteins and sequences involved in the structure of the nervous system, for example at

synapses, are thus assigned to their function with the aid of genome, sequence and domain

analyses (as already practised on other topics).

However, neurobiological processes can also be viewed using a wide variety of other

bioinformatics techniques. The structure of important receptors involved can be modelled,

e.g. the seven transmembrane helices that build up a GPCR receptor, as well as other

important activatory (e.g. glutamate) and inhibitory receptors (e.g. glycine). At the next

level, neuronal networks can be represented by semiquantitative simulations, but also

receptor excitations, for example by differential equations. Individual circuit diagrams can

be recreated in the computer, for example for memory (Rolls 2013a) or for the whole net­

work of the cordworm C. elegans (Connectome C. elegans). For the hippocampus, there

are already ideas about how it recognizes and separates patterns (Rolls 2013b). There is

even a quantitative theory of the function of individual layers in the hippocampus (Rolls

2013a). Higher processes can also be modelled with the help of computers (Markram et al.

2015). In particular, however, omics studies again provide an overview of the brain (espe­

cially, but not only, of humans, e.g. an atlas of gene expression in the human brain;

Hawrylycz et al. 2015). Much is also taking place on the model organism mouse.

An overview of the pathology is equally important. Bioinformatics can help here, for

example, to better understand the regeneration processes in the old and young brain in

stroke using transcriptome data (Buga et al. 2012) or, for example, to decipher regenera­

tion in the hippocampus using various data via statistical analyses, whereby erythropoietin

apparently has a supporting effect (Hassouna et al. 2016). Meanwhile, there are a whole

bunch of ways to boost memory performance using growth factors, stem cells, pharmaceu­

ticals, or even memory training. But a clinical breakthrough, for example in Alzheimer’s

disease, will still require a lot of work and studies (Schneider et al. 2020).

15  How Is Our Own Extremely Powerful Brain Constructed?