Recent Progress on Methods for Estimating and Updating Large Phylogenies

With the increased availability of sequence data and even of fully sequenced and assembled genomes, phylogeny estimation of very large trees (even of hundreds of thousands of sequences) is now a goal for some biologists. Yet, the construction of these phylogenies is a complex pipeline presenting analytical and computational challenges, especially when the number of sequences is very large. In the last few years, new methods have been developed that aim to enable highly accurate phylogeny estimations on these large datasets, including divide-and-conquer techniques for multiple sequence alignment and/or tree estimation, methods that can estimate species trees from multi-locus datasets while addressing heterogeneity due to biological processes (e.g., incomplete lineage sorting and gene duplication and loss), and methods to add sequences into large gene trees or species trees. Here we present some of these recent advances and discuss opportunities for future improvements.

Novel Neutralizing Epitope of PEDV S1 Protein Identified by IgM Monoclonal Antibody

Porcine epidemic diarrhea virus (PEDV) causes devastating enteric disease that inflicts huge economic damage on the swine industry worldwide. A safe and highly effective PEDV vaccine that contains only the virus-neutralizing epitopes (not enhancing epitope), as well as a ready-to-use PEDV neutralizing antibody for the passive immunization of PEDV vulnerable piglets (during the first week of life) are needed, particularly for PEDV-endemic farms. In this study, we generated monoclonal antibodies (mAbs) to the recombinant S1 domain of PEDV spike (S) protein and tested their PEDV neutralizing activity by CPE-reduction assay. The mAb secreted by one hybrodoma clone (A3), that also bound to the native S1 counterpart from PEDV-infected cells (tested by combined co-immunoprecipitation and Western blotting), neutralized PEDV infectivity. Epitope of the neutralizing mAb (mAbA3) locates in the S1A subdomain of the spike protein, as identified by phage mimotope search and multiple sequence alignment, and peptide binding-ELISA. The newly identified epitope is shared by PEDV G1 and G2 strains and other alphacoronaviruses. In summary, mAbA3 may be useful as a ready-to-use antibody for passive immunization of PEDV-susceptible piglets, while the novel neutralizing epitope, together with other, previously known protective epitopes, have potential as an immunogenic cocktail for a safe, next-generation PEDV vaccine.

Harnessing protein folding neural networks for peptide–protein docking

Highly accurate protein structure predictions by deep neural networks such as AlphaFold2 and RoseTTAFold have tremendous impact on structural biology and beyond. Here, we show that, although these deep learning approaches have originally been developed for the in silico folding of protein monomers, AlphaFold2 also enables quick and accurate modeling of peptide–protein interactions. Our simple implementation of AlphaFold2 generates peptide–protein complex models without requiring multiple sequence alignment information for the peptide partner, and can handle binding-induced conformational changes of the receptor. We explore what AlphaFold2 has memorized and learned, and describe specific examples that highlight differences compared to state-of-the-art peptide docking protocol PIPER-FlexPepDock. These results show that AlphaFold2 holds great promise for providing structural insight into a wide range of peptide–protein complexes, serving as a starting point for the detailed characterization and manipulation of these interactions.

Characterization of Subtilosin Gene In Wild Type Bacillus Spp. And Possible Physiological Role

In a designed study to screen for antimicrobial exhibiting bacteria using molecular aspects, Bacillus species were considered to investigate antibiotic biosynthesis genes. 28 bacterial strains and 3 induced mutants were screened for the presence of subtilosin gene (sbo) and subtilosin through PCR and Mass spectrometry respectively. Sbo gene was detected in 16 out of 28 Bacillus strains. The results from gene sequences deliberated by multiple sequence alignments revealed high-level homology to the sequences of the sbo-alb gene locus of B. subtilis 168 and the other limited reported strains. Hence, this report provided additional strains to support the idea of subtilosin gene predominance amongst Bacillus strains isolated from environment and to find different species containing homologous genes, furthermore the utilization of its conserved region as a means of identifying Bacillus spp. that produce subtilosin. This is the first report to confirm the detection of subtilosin production from B. amyloliquefaciens.

Genetic variability and population structure of Leucinodes orbonalis (Guenée), a severe insect pest of brinjal in India

Aim: The present study was undertaken to provide valuable insights regarding population genetic structure of Leucinodes orbonalis from diverse agro-ecologies of India. Methodology: Molecular characterization of L. orbonalis populations collected from five major agro-climatic zones of India was carried out using mitochondrial cytochrome oxidase I (COI) gene. Collected specimens were subjected to DNA extractions, partial PCR amplification and sequencing of the target gene, and multiple sequence alignments. Results: The results showed very less diversity in the nucleotide positions of the COI sequences of 79 studied specimens, with a low number of segregating sites (30), nucleotide diversity (0.00438) and overall mean genetic distance (0.004 ± 0.001). The significant negative values of neutrality tests and unimodal mismatch distribution supported the demographic expansion theory in Indian L. orbonalis. Analysis of the molecular variance revealed that 93.13% of the genetic variation was within populations, and the variation among populations was only 6.87%. The pairwise genetic differentiation was also found to be low to moderate between most of the populations. Multiple haplotypes were recorded from all the populations, and both neighbour-joining tree as well as the haplotype network showed that clustering of the haplotypes was independent of the geographical location. Interpretation: Thus, it can be inferred that Indian populations of L. orbonalis have very low genetic variation levels concerning the COI gene. There is a possible occurrence of stable inherited gene flow among populations, thereby reducing genetic variation in India.

Molecular characterization and phylogenetic analysis of bovine ephemeral fever viruses in Khuzestan province of Iran in 2018 and 2020

Background Bovine ephemeral fever (BEF) is an arthropod-borne viral disease caused by the BEF virus (BEFV). This single-stranded RNA virus that affects cattle and water buffalo is endemic in tropical and subtropical regions including Iran. While BEF is a major disease of cattle in Iran, information regarding its agent, molecular characterization, and circulating viruses are highly limited. The current study aimed to, firstly, determine the genetic and antigenic characteristics of BEFV strains in Khuzestan province in Southwest of Iran in 2018 and 2020 and, secondly, to compare them with strains obtained from other areas. Results By phylogenetic analysis based on the Glycoprotein gene, BEFV strains were divided into four clusters of Middle East, East Asia, South Africa, and Australia; in which the 2018 and 2020 Iranian BEFV strains were grouped in the Middle East cluster with the Turkish, Indian, and Israeli strains. Depending on the chronology and geographical area, the outbreaks of Turkey (2020), Iran (2018 and 2020), and India (2018 and 2019) are proposed to be related. These BEFVs had the highest identity matrix and the lowest evolutionary distance among the studied strains. Multiple sequence alignment of G1, G2, and G3 antigenic sites showed that these neutralizing epitopes are highly conserved among the strains of the Middle East cluster; however, the strains previously identified in Iran differed in three amino acids placed in G1 and G2 epitopes. Conclusion The findings revealed that BEFVs circulating in the Middle East are closely related phylogenetically and geographically. They also have similar antigenic structures; therefore, developing a vaccine based on these strains can be effective for controlling BEF in the Middle East.

NeoRdRp: A comprehensive dataset for identifying RNA-dependent RNA polymerase of various RNA viruses from metatranscriptomic data

RNA viruses are distributed in various environments, and most RNA viruses have been recently identified by metatranscriptome sequencing. However, due to the high nucleotide diversity of RNA viruses, it is still challenging to identify their sequences. Therefore, this study generated a dataset of RNA-dependent RNA polymerase (RdRp) domains essential for all RNA viruses belonging to Orthornavirae. Also, the collected genes with RdRp domains from various RNA viruses were clustered by amino acid sequence similarity. For each cluster, a multiple sequence alignment was generated, and a hidden Markov model (HMM) profile was created if the number of sequences was greater than five. Using the 1,467 HMM profiles, we detected RdRp domains in the RefSeq RNA virus sequences, combined the hit sequences with the RdRp domains, and reconstructed the HMM profiles. As a result, 2,234 HMM profiles were generated from 12,316 RdRp domain sequences, and the dataset was named NeoRdRp. Additionally, using the UniProt dataset, we confirmed that almost all NeoRdRp HMM profiles could specifically detect RdRps in Orthornavirae. Furthermore, we compared the NeoRdRp dataset with two previously reported RNA virus detection methods to detect RNA virus sequences from metatranscriptome sequencing data. Our methods can identify most of the RNA viruses in the datasets; however, some RNA viruses were not detected, similar to the other two methods. The NeoRdRp can be improved by repeatedly adding new RdRp sequences and can be expected to be widely applied as a system for detecting various RNA viruses from metatranscriptome data.

Embeddings from protein language models predict conservation and variant effects

The emergence of SARS-CoV-2 variants stressed the demand for tools allowing to interpret the effect of single amino acid variants (SAVs) on protein function. While Deep Mutational Scanning (DMS) sets continue to expand our understanding of the mutational landscape of single proteins, the results continue to challenge analyses. Protein Language Models (pLMs) use the latest deep learning (DL) algorithms to leverage growing databases of protein sequences. These methods learn to predict missing or masked amino acids from the context of entire sequence regions. Here, we used pLM representations (embeddings) to predict sequence conservation and SAV effects without multiple sequence alignments (MSAs). Embeddings alone predicted residue conservation almost as accurately from single sequences as ConSeq using MSAs (two-state Matthews Correlation Coefficient—MCC—for ProtT5 embeddings of 0.596 ± 0.006 vs. 0.608 ± 0.006 for ConSeq). Inputting the conservation prediction along with BLOSUM62 substitution scores and pLM mask reconstruction probabilities into a simplistic logistic regression (LR) ensemble for Variant Effect Score Prediction without Alignments (VESPA) predicted SAV effect magnitude without any optimization on DMS data. Comparing predictions for a standard set of 39 DMS experiments to other methods (incl. ESM-1v, DeepSequence, and GEMME) revealed our approach as competitive with the state-of-the-art (SOTA) methods using MSA input. No method outperformed all others, neither consistently nor statistically significantly, independently of the performance measure applied (Spearman and Pearson correlation). Finally, we investigated binary effect predictions on DMS experiments for four human proteins. Overall, embedding-based methods have become competitive with methods relying on MSAs for SAV effect prediction at a fraction of the costs in computing/energy. Our method predicted SAV effects for the entire human proteome (~ 20 k proteins) within 40 min on one Nvidia Quadro RTX 8000. All methods and data sets are freely available for local and online execution through bioembeddings.com, https://github.com/Rostlab/VESPA, and PredictProtein.