Dataset Information

Feature-extraction and analysis based on spatial distribution of amino acids for SARS-CoV-2 Protein sequences.

ABSTRACT:

Background and objective

The world is currently facing a global emergency due to COVID-19, which requires immediate strategies to strengthen healthcare facilities and prevent further deaths. To achieve effective remedies and solutions, research on different aspects, including the genomic and proteomic level characterizations of SARS-CoV-2, are critical. In this work, the spatial representation/composition and distribution frequency of 20 amino acids across the primary protein sequences of SARS-CoV-2 were examined according to different parameters.

Method

To identify the spatial distribution of amino acids over the primary protein sequences of SARS-CoV-2, the Hurst exponent and Shannon entropy were applied as parameters to fetch the autocorrelation and amount of information over the spatial representations. The frequency distribution of each amino acid over the protein sequences was also evaluated. In the case of a one-dimensional sequence, the Hurst exponent (HE) was utilized due to its linear relationship with the fractal dimension (D), i.e. D+HE=2, to characterize fractality. Moreover, binary Shannon entropy was considered to measure the uncertainty in a binary sequence then further applied to calculate amino acid conservation in the primary protein sequences.

Results and conclusion

Fourteen (14) SARS-CoV protein sequences were evaluated and compared with 105 SARS-CoV-2 proteins. The simulation results demonstrate the differences in the collected information about the amino acid spatial distribution in the SARS-CoV-2 and SARS-CoV proteins, enabling researchers to distinguish between the two types of CoV. The spatial arrangement of amino acids also reveals similarities and dissimilarities among the important structural proteins, E, M, N and S, which is pivotal to establish an evolutionary tree with other CoV strains.

SUBMITTER: Rout RK

PROVIDER: S-EPMC8577876 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BACKGROUND: Many learning approaches to predicting RNA-binding residues in a protein sequence construct a non-redundant training dataset based on the sequence similarity. The sequence similarity-based method either takes a whole sequence or discards it for a training dataset. However, similar sequences or even identical sequences can have different interaction sites depending on their interaction partners, and this information is lost when the sequences are removed. Furthermore, a training dataset constructed by the sequence similarity-based method may contain redundant data when the remaining sequence contains similar subsequences within the sequence. In addition to the problem with the training dataset, most approaches do not consider the interacting partner (i.e., RNA) of a protein when they predict RNA-binding amino acids. Thus, they always predict the same RNA-binding sites for a given protein sequence even if the protein binds to different RNA molecules. RESULTS: We developed a feature vector-based method that removes data redundancy for a non-redundant training dataset. The feature vector-based method constructed a larger training dataset than the standard sequence similarity-based method, yet the dataset contained no redundant data. We identified effective features of protein and RNA (the interaction propensity of amino acid triplets, global features of the protein sequence, and RNA feature) for predicting RNA-binding residues. Using the method and features, we built a support vector machine (SVM) model that predicted RNA-binding residues in a protein sequence. Our SVM model showed an accuracy of 84.2%, an F-measure of 76.1%, and a correlation coefficient of 0.41 with 5-fold cross validation on a non-redundant dataset from 3,149 protein-RNA interacting pairs. In an independent test dataset that does not include the 3,149 pairs and were not used in training the SVM model, it achieved an accuracy of 90.3%, an F-measure of 72.8%, and a correlation coefficient of 0.24. Comparison with other methods on the same datasets demonstrated that our model was better than the others. CONCLUSIONS: The feature vector-based redundancy reduction method is powerful for constructing a non-redundant training dataset for a learning model since it generates a larger dataset with non-redundant data than the standard sequence similarity-based method. Including the features of both RNA and protein sequences in a feature vector results in better performance than using the protein features only when predicting the RNA-binding residues in a protein sequence.

Project description:Since its first emergence in December 2019, the world has witnessed the eruption of mutations in the SARS-CoV-2 genome that have led to increased viral transmissibility and pathogenicity due to sustained local viral transmission. Zooanthroponotic and zoonotic transmissions have further raised concerns as they could result in the emergence of viral variants with a novel antigenicity and transmissibility that could jeopardize the vaccine efficacy. To understand the viral evolution during such transmissions, 1016 whole-genome sequences (deposited in GISAID as of March 7, 2022) (from 18 countries) corresponding to mink, cat, deer, dog, hyena, tiger, lion, gorilla, Syrian hamster, leopard cat, fishing cat, bear cat, coati, ferret, snow leopard and green monkey have been analysed here. Intriguingly, phyloproteome analysis indicate that Nsp2:R218C, Nsp2:D268-(deletion), Spike:D614G, Nsp12:P323L, Nsp2:A192V, ORF3a protein:Q57H, N protein:R203K and N protein:G204R/L, Spike:A222V, ORF10 protein:V30L and N protein:A220V are moderate or high recurring and clade decisive mutations, leading to 6 primary clades during the early stage of pandemic. Most interestingly, the human evolved delta variant having a combination of 26 (clade decisive) mutations defines the seventh clade and transmits to non-human hosts across the globe without exhibiting any country-specific mutation(s). Nonetheless, Spike:D614G and Nsp12:P323L together with (i)N protein:R203K,N protein:G204R/L,Spike:V70-, Spike:H69-, Nsp12:T739I, and Nsp1:M85-, (ii)Nsp2:A192V, Nsp3:D178Y, (iii)Nsp2:T85I, N protein:P67S and ORF3a protein:Q57H and (iv)Spike:A222V, ORF10 protein:V30L, N protein:A220V and Spike:F486I are specific to Denmark, Netherlands, USA and Latvia respectively and, (v)Nsp2:D268- and Nsp13:R292C that are devoid of Spike:D614G and Nsp12:P323L is specific to Netherlands. SARS-CoV-2 variants consisting of these mutations are also seen in the human SARS-CoV-2 sequences from the same country. Independent country-specific SARS-CoV-2 variant evolution further indicates distinct epidemiological dynamics during zooanthroponotic and zoonotic transmissions. Thus, the results presented here indicate the need for the surveillance of viral evolution in non-human hosts also during the future pandemic.