Search

In this study, we analyze the role of different structural variants of proteins in the speciation processes. We separate human and mouse proteomes (taken as a reference) into three previously defined variants of disorder: ordered proteins (ORDPs), structured proteins with intrinsically disordered protein regions (IDPRs), and intrinsically disordered proteins (IDPs). Then, using the representation we call here Forsdyke plot, we study the correlation of DNA divergence with the corresponding protein (phenotypic) divergence in the three variants, comparing human and mouse coding sequences with their homologs from 26 eukaryotes. The parameters of the correlation are related to the speciation process. We find that the three variants of disordered proteins are differently related to the speciation process. Specifically, IDPs phenotypically diverge earlier than ORDPs and IDPRs. ORDPs diverge later but are phenotypically more reactive to nucleotide mutations than IDPRs and IDPs. Finally, IDPRs appear to diverge phenotypically later than IDPs, like ORDPs, but they are prone to accept mutations with rates that are similar to those of IDPs. We conclude that IDPs are involved in the early stages of the speciation process, whereas mutations in ORDPs, once speciation is initiated, accelerate phenotypic divergence.

In this study, we perform a systematic analysis of evolutionary forces (i.e., mutational bias and natural selection) that shape the codon usage bias of human genes encoding proteins characterized by different flavors of intrinsic disorder. Well-structured proteins are expected to be more under control by purifying natural selection than intrinsically disordered proteins because one or few mutations (even synonymous) in the genes can result in a protein that no longer folds correctly. On the contrary, intrinsically disordered proteins are thought to evolve more rapidly than well-folded proteins, due to a relaxed purifying natural selection and an increased role of mutational bias. Using different bioinformatic tools, we find evidence that codon usage in IDPs is not only affected by a basic mutational bias, but it is also more selectively constrained than the rest of the human proteome. We speculate that intrinsically disordered proteins have not only a high tolerance to mutations but also a selective propensity to preserve their structural disorder under physiological conditions. Additionally, we confirm not only that intrinsically disordered proteins are preferentially encoded by GC-rich genes, but also that they are characterized by the highest fraction of CpG sites in the sequences, implying a higher susceptibility to methylation resulting in C-T transition mutations. Overall, our results corroborate the essential role of intrinsic disorder for the evolutionary adaptability and evolvability of proteins, offering new insight about protein evolution not only in terms of functional properties and roles in diseases but also in terms of evolutionary forces they are subjected to.

The outbreak of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) has caused an unprecedented pandemic. Since the first sequenced whole-genome of SARS-CoV-2 on January 2020, the identification of its genetic variants has become crucial in tracking and evaluating their spread across the globe.In this study, we compared 134,905 SARS-CoV-2 genomes isolated from all affected countries since the outbreak of this novel coronavirus with the first sequenced genome in Wuhan, China to quantify the evolutionary divergence of SARS-CoV-2. Thus, we compared the codon usage patterns of SARS-CoV-2 genes encoding the membrane protein (M), envelope (E), spike surface glycoprotein (S), nucleoprotein (N), RNA-dependent RNA polymerase (RdRp). The polyproteins ORF1a and ORF1b were examined separately.We found that SARS-CoV-2 tends to diverge over time by accumulating mutations on its genome and, specifically, on the sequences encoding proteins N and S. Interestingly, different patterns of codon usage were observed among these genes. Genes S and N tend to use a narrower set of synonymous codons that are better optimized to the human host. Conversely, genes E and M consistently use the broader set of synonymous codons, which does not vary in respect to the reference genome. CAI and SiD time evolutions show a tendency to decrease that emerge for most genes. Forsdyke plots are used to study the nature of mutations and they show a rapid evolutionary divergence of each gene, due to the low values of x-intercepets.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which first occurred in Wuhan (China) in December 2019, is a novel virus that causes a severe acute respiratory disease. The virus spike glycoproteins and nucleocapsid proteins are the main targets for the development of vaccines and antiviral drugs, to control the disease spread. We herein study the structural order-disorder propensity and the rates of evolution of these two proteins to characterize their B cell and T cell epitopes, previously suggested to contribute to immune response caused by SARS-CoV-2 infections. We first analyzed the rates of evolution along the sequences of spike and nucleocapsid proteins in relation to the spatial locations of their epitopes. For this purpose, we compared orthologs from seven human coronaviruses: SARS-CoV-2, SARS-CoV, MERS-CoV, HCoV-229E, HCoV-OC43, HCoV-NL63, and HCoV-HKU1. We then focus on the local, structural order-disorder propensities of the protein regions where the SARS-CoV-2 epitopes are located. We show that the vast majority of nucleocapsid protein epitopes overlap the RNA-binding and dimerization domains and some of them are characterized by low rates of evolutions. Similarly, spike protein epitopes are preferentially located in regions that are predicted to be ordered and well-conserved, in correspondence of the heptad repeats 1 and 2. Interestingly, both the receptor-binding motif to ACE2 and the fusion peptide of spike protein are characterized by high rates of evolution, probably to overcome host immunity. In conclusion, our results provide evidence for conserved epitopes that may help to develop long-lasting, broad-spectrum SARS-CoV-2 vaccines.

We study the correlation between the codon usage bias of genetic sequences and the network features of protein-protein interaction (PPI) in bacterial species. We use PCA techniques in the space of codon bias indices to show that genes with similar patterns of codon usage have a significantly higher probability that their encoded proteins are functionally connected and interacting. Importantly, this signal emerges when multiple aspects of codon bias are taken into account at the same time. The present study extends our previous observations on E. coli over a wide set of 34 bacteria. These findings could allow for future investigations on the possible effects of codon bias on the topology of the PPI network, with the aim of improving existing bioinformatics methods for predicting protein interactions.

Severe acute respiratory syndrome coronavirus 2 (2019-nCoV), which first broke out in Wuhan (China) in December of 2019, causes a severe acute respiratory illness with a mortality ranging from 3% to 6%. To better understand the evolution of the newly emerging 2019-nCoV, in this paper, we analyze the codon usage pattern of 2019-nCoV. For this purpose, we compare the codon usage of 2019-nCoV with that of other 30 viruses belonging to the subfamily of orthocoronavirinae. We found that 2019-nCoV shows a rich composition of AT nucleotides that strongly influences its codon usage, which appears to be not optimized to human host. Then, we study the evolutionary pressures influencing the codon usage and evolutionary rates of the sequences of five conserved genes that encode the corresponding proteins (viral replicase, spike, envelope, membrane and nucleocapsid) characteristic of coronaviruses. We found different patterns of both mutational bias and nature selection that affect the codon usage of these genes at different extents. Moreover, we show that the two integral membrane proteins proteins (matrix and envelope) tend to evolve slowly by accumulating nucleotide mutations on their genes. Conversely, genes encoding nucleocapsid (N), viral replicase and spike proteins are important targets for the development of vaccines and antiviral drugs, tend to evolve faster as compared to other ones. Taken together, our results suggest that the higher evolutionary rate observed for these two genes could represent a major barrier in the development of antiviral therapeutics 2019-nCoV.

One of the most debated topics in Evolutionary Biology concerns Low Complexity Regions of P. falciparum, the causative agent of the most virulent and deadly form of human malaria. In this work, we analysed the proteome of 22 plasmodium species including P. falciparum. SEG predicts that proteins containing Low Complexity Regions turn out to be longer than those which are predicted to be completely complex (without Low Complexity Regions). Moreover, using some well-known bioinformatics tools such as the Effective Number of Codons, the Pr2 and a new index that we have called SPI, we have noticed how proteins that embed Low Complexity Regions are under lower selective pressure than those that do not present this type of locus. By applying the Relative Synonymous Codon Usage and other tools developed ad hoc for this study, we note, instead, how the Low Complexity Regions appear to have a non-neutral codon bias with respect to the host proteins.

In a recent study, we have introduced an operational classification of the human proteome in three variants of disorder: ordered proteins (ORDPs), structured proteins with intrinsically disordered protein regions (IDPRs), intrinsically disordered proteins (IDPs). That classification was useful in functionally separating IDPRs from IDPs, which up until now have been generally considered as a whole. In this study, we corroborate this distinction by considering different physical-chemical and structural properties. Both ORDPs and IDPRs are enriched in order-promoting amino acids, whereas only IDPs show an enrichment in disordered-promoting amino acids. Consistently, ORDPs and IDPRs are preferentially located in the ordered phase of the charge-hydropathy plot, whereas IDPs are widespread over the disordered phase. We introduce the mean packing - mean pairwise energy (MP-MPE) plane to structurally characterize these variants even in the absence of a structural model. As expected for well-packed proteins, a negative linear correlation is observed between MP and MPE for ORDPs and IDPRs, whereas IDPs break this linear dependence. Finally, we find that IDPs have a more extended conformation as measured by the scaling law between the radius of gyration and the length of these proteins, and accordingly they have higher solubility and accessible surface area than ORDPs and IDPRs. Overall, our results confirm the relevance of our operational separation of IDPRs from IDPs and provide further validation of our criteria to separate IDPs from the rest of human proteome.

Taking the human genome as a case of study, we propose a new classification of codons based only on two genomic information. We use the relative synonymous codon usage (RSCU) as a measure of non-uniform usage of synonymous codons. Similarly, we introduce here the relative gene frequencies of cognate tRNAs (RGFCt) to quantify the non-uniform availability of cognate tRNAs in each family of synonymous codons. Using these two quantities, we define two general groups of codons: non-frustrated codons, whose usage in the coding sequences is in proportion to the expected cognate tRNA levels, and frustrated codons, which do not satisfy this proportionality. With this decoding for every codon, we defined the Codon Frustration Index (CFI) as the net frustration of a gene, normalized for its length. Notably, we find that CFI correlates very well with other independent measures of CUB and a high content of non-frustrated codons increases both translation efficiency and mRNA stability. Finally, we show that genes with either a high content of frustrated or of non-frustrated codons are differentially enriched in specific functional classes that typically comprise nucleic acid binding proteins, mRNA processing factors, RNA helicase, and in several transcription factors.

In this study, we perform a systematic analysis of evolutionary forces (i.e., mutational bias and natural selection) that shape the codon usage bias of human genes encoding for different structural and functional variants of proteins. Well-structured proteins are expected to be more under control by natural selection than intrinsically disordered proteins because one or few mutations (even synonymous) in the genes can result in a protein that no longer folds correctly. On the contrary, intrinsically disordered proteins are generally thought to evolve more rapidly than well-folded proteins, primarily attributed to relaxed purifying natural selection due to the lack of structural constraints. Using different genetic tools, we find compelling evidence that intrinsically disordered proteins are the variant of human proteins on which both mutational bias and natural selection act more effectively, corroborating their essential role for evolutionary adaptability and protein evolvability. We speculate that intrinsically disordered proteins have a high tolerance to mutations (both neutral and adaptive) but also a selective propensity to preserve their structural disorder, i.e., flexibility and conformational dynamics under physiological conditions. Additionally, we confirm not only that intrinsically disordered proteins are preferentially encoded by GC-rich genes, but also that they are characterized by the highest fraction of CpG-sites in the sequences, implying a higher susceptibility to methylation resulting in C-T transition mutations. Our results provide new insight about protein evolution and human genetic diseases identifying intrinsically disordered proteins as reservoirs for evolutionary innovations.

The codon usage bias is the well-known phenomenon of an unequal use of synonymous codons in coding DNA. These patterns reflect the action of weak selection working at the molecular level and allows to quantify the effects of natural selection, which tend to increase the fitness of the organisms. The prevailing hypothesis to explain the origin of codons usage bias is the selection-mutation-drift theory, according to which it results from a balance between the natural selection favoring optimal codons and combined action of random mutations and genetic drift which allow the persistence of nonoptimal codons. The main focus of this study is to quantify the extent of evolutionary pressures shaping the human genome. We found distinct patterns of mutational bias and natural selection in the human genes, depending on the structural properties of the encoded proteins (e.g. well-structured proteins, proteins with a long disordered segment embedded in a folded structure, or mostly unfolded proteins). Intrinsically disordered proteins are generally thought to evolve more rapidly, largely attributed to relaxed purifying selection due to the lack of structural constraint. Interestingly we observed that mostly unstructured proteins are not only affected by a basic mutational bias as the structured ones but are under a specific selective pressure underlining the important role of these proteins during evolution being freer to accept mutations, both neutral and selective. Our results provide new insight into understanding general laws and unknown aspects of protein evolution and they could be very useful in protein search and design.

We propose a new, sequence-only, classification of intrinsically disordered human proteins which is based on two parameters: dr, the percentage of disordered residues, and Ld, the length of the longest disordered segment in the sequence. Depending on dr and Ld, we distinguish five variants: i)ordered proteins(ORDs); ii)not disordered proteins(NDPsj; (iii)proteins with intrinsically disordered regions(PDRs); iv)intrinsically disordered proteins(IDPs) and v)proteins with fragmenteddisorder (FRAGs). PDRs have been considered in the general category of intrinsically disordered proteins for a long time. We show that PDRs are closer to globular, ordered proteins (ORDs and NDPs) than to disordered ones (IDPs), both in amino acid composition and functionally. Moreover, NDPs and PDRs are uniformly spread over several functional protein classes, whereas IDPs are concentrated only on two, namelynucleic acid binding proteinsandtranscription factors, which are just a subset of the functions that are commonly associated with protein intrinsic disorder. As a conclusion, PDRs and IDPs should be considered, in future classifications, as distinct variants of disordered proteins, with different physical-chemical properties and functional spectra.