Community

LinkedIn post

https://www.linkedin.com/feed/update/urn:li:activity:7365320270401757184

One configuration

One of the main questions about the Q-mosaic property is the characteristics of the mosaic itself. That is, is there only one configuration possible? By configuration, I mean here the size (number of residues) and shape of the clusters that conform the mosaic.

Well, the stochastic simulations that are part of the project suggest that only one configuration explains the experimental results. That is, varying the cluster size or shape directly affects the Q-n curve obtained, and we must remember that only one Q-n curve is found in nature, so that only one configuration apparently exists. According to the stochastic simulations, the configuration (or at least a good approximation to it) is of 8 residues per cluster, with Shape I (see Github for details). That does not mean that all clusters have exactly the same size and shape, rather we are talking about this specific size and shape with some random noise added to it (although not much random noise, since R^2 is actually very high). Visualization of examples (see the Repository) helps confirm this. 

The mosaic is actually a patchy pattern of residues in the 3D structure of proteins, where each patch contains residues of a specific chemical type (basic, acidic, polar, hydrophobic). What is relevant here is that not only a patchy pattern appears in apparently all proteins, but that the morphology of the patchs are nearly identical in all of them. Alternatively stated, only one configuration of the mosaic seems present in polypeptides.

We are assuming that conclusions from analyzing the whole PDB database are valid, or at least from the analysis of those entries obtained via X-ray diffraction, which are currently more than 160,000. Obviously, there are dozens of millions of proteins whose structure is still unknown, so making statements about the totallity of structures must be done with caution. Nevertheless, >160,000 is already a considerable figure, especially taking into account that those entries include proteins from all kind of living organisms, from humans, to plants, fungi or bacteria, and also viral proteins.

Some questions remain, such as why that particular configuration and not other? Is that configuration related to protein structural stability? Or maybe is it related to protein functioning? Perhaps, is it selected via natural selection (so that all new proteins that appear in evolution must match the same mosaic configuration as the already existing ones)? Only the involvement of the scientific community will help solving these puzzles. Become an early adopter ☀️

The importance of chemistry

It is somewhat surprising that amino acids in proteins are clustered according to their chemical type. There are 20 different standard amino acids in proteins (and some more, although less common), each having their specific physical characteristics. However, for the Q-mosaic, only the chemical type of the amino acid is important type (polar, hydrophobic, basic, acidic).

Once, I heard that chemistry is but an abstraction of physics. In this case, the specific features of each of the 20 amino acids could be considered their physical features, whilst their grouping in four families is more of a chemistry thing. Concerning the Q-mosaic, it's as if nature had done precisely this, in other words, what is relevant --at least as far as the Q-mosaic is concerned- is their chemical type. And hence the importance of chemistry. Grouping amino acids in four groups (polar, hydrophobic, basic, acidic) is a human classification, but actually it seems that polypeptide structure has chosen the same abstraction in the configuration of the mosaic.

We also need to remember that the mosaic is not all about the chemical type of the clusters, but also about the size and shape of those clusters. In other words, the property is not chemical but rather chemical-geometrical, with well defined rules that delimit the chemical types and what the morphology of the clusters is like.

In conclusion, chemistry (or alternatively stated, abstraction) is apparently relevant for protein structure, at least for the Q-mosaic property!!