Tsakris, A. | bioRxiv Science

bioRxiv2023

Paired associated SARS-CoV-2 spike variable positions: a network analysis approach to emerging variants

Amino acids in variable positions of proteins may be correlated, with potential structural and functional implications. Here, we apply exact tests of independence in R x C contingency tables to examine noise-free associations between variable positions of the SARS-CoV-2 spike protein, using as a paradigm sequences from Greece deposited in GISAID (N=6,683/1,078 full-length) for the period February 29, 2020 to April 26, 2021 that essentially covers the first three pandemic waves. We examine the fate and complexity of these associations by network analysis, using associated positions (exact p[≤]0.001 and Average Product Correction [≥]2) as links and the corresponding positions as nodes . We found a temporal linear increase of positional differences and a gradual expansion of the number of position associations over time, represented by a temporally evolving intricate web, resulting in a non-random complex network of 69 nodes and 252 links. Overconnected nodes corresponded to the most adapted variant positions in the population, suggesting a direct relation between network degree and position functional importance. Modular analysis revealed 25 k-cliques comprising three to 11 nodes. At different k- clique resolutions, one to four communities were formed, capturing epistatic associations of circulating variants (Alpha, Beta, B.1.1.318), but also Delta, which dominated the evolutionary landscape later in the pandemic. Cliques of aminoacidic positional associations tended to occur in single sequences, enabling the recognition of epistatic positions in real-world virus populations. Our findings provide a novel way of understanding epistatic relationships in viral proteins with potential applications in the design of virus control procedures. ImportancePaired positional associations of adapted amino acids in virus proteins may provide new insights for understanding virus evolution and variant formation. We investigated potential intramolecular relationships between variable SARS-CoV-2 spike positions by exact tests of independence in R x C contingency tables, having applied Average Product Correction (APC) to eliminate background noise. Associated positions (exact p[≤]0.001 and APC[≥]2) formed a non- random, epistatic network of 25 cliques and 1-4 communities at different clique resolutions, revealing evolutionary ties between variable positions of circulating variants, and a predictive potential of previously unknown network positions. Cliques of different sizes represented theoretical combinations of changing residues in sequence space, allowing the identification of significant aminoacidic combinations in single sequences of real-world populations. Our analytic approach that links network structural aspects to mutational aminoacidic combinations in the spike sequence population offers a novel way to understand virus epidemiology and evolution.

microbiology↗

bioRxiv2020

Pixel-based machine learning and image reconstitution for dot-ELISA pathogen serodiagnosis

Serological methods serve as a direct or indirect means of pathogen infection diagnosis in plant and animal species, including humans. Dot-ELISA (DE) is an inexpensive and sensitive, solid-state version of the microplate enzyme-linked immunosorbent assay, with a broad range of applications in epidemiology. Yet, its applicability is limited by uncertainties in the qualitative output of the assay due to overlapping dot colorations of positive and negative samples, stemming mainly from the inherent color discrimination thresholds of the human eye. Here, we report a novel approach for unambiguous DE output evaluation by applying machine learning-based pattern recognition of image pixels of the blot using an impartial predictive model rather than human judgment. Supervised machine learning was used to train a classifier algorithm through a built multivariate logistic regression model based on the RGB ("Red", "Green", "Blue") pixel attributes of a scanned DE output of samples of known infection status to a model pathogen (Lettuce big-vein associated virus). Based on the trained and cross-validated algorithm, pixel probabilities of unknown samples could be predicted in scanned DE output images which would then be reconstituted by pixels having probabilities above a cutoff that may be selected at will to yield desirable false positive and false negative rates depending on the question at hand, thus allowing for proper dot classification of positive and negative samples and, hence, accurate diagnosis. Potential improvements and diagnostic applications of the proposed versatile method that translates unique pathogen antigens to the universal basic color language are discussed.

microbiology↗