Abstract: Glycosaminoglycans (GAGs) are linear polysaccharides. In proteoglycans (PGs), they are attached to a core protein. GAGs and PGs can be found as free molecules, associated with the extracellular matrix or expressed on the cell membrane. They play a role in the regulation of a wide array of physiological and pathological processes by binding to different proteins, thus modulating their structure and function, and their concentration and availability in the microenvironment. Unfortunately, the enormous structural diversity of GAGs/PGs has hampered the development of dedicated analytical technologies and experimental models. Similarly, computational approaches (in particular, molecular modeling, docking and dynamics simulations) have not been fully exploited in glycobiology, despite their potential to demystify the complexity of GAGs/PGs at a structural and functional level. Here, we review the state-of-the art of computational approaches to studying GAGs/PGs with the aim of pointing out the "bitter" and "sweet" aspects of this field of research. Furthermore, we attempt to bridge the gap between bioinformatics and glycobiology, which have so far been kept apart by conceptual and technical differences. For this purpose, we provide computational scientists and glycobiologists with the fundamentals of these two fields of research, with the aim of creating opportunities for their combined exploitation, and thereby contributing to a substantial improvement in scientific knowledge.

Abstract: With the advancement of the Covid-19 pandemic, this work aims to find molecules that can inhibit the attraction between the Spike proteins of the SARS-COV-2 virus and human ACE2. The results of molecular docking positioned four molecules at the interaction site Tyr-491(Spike)-Glu-37(ACE2) and one at the site Gly-488(Spike)-Lys-353(ACE2). The QTAIM and IQA data showed that the 1629 molecule had a significant inhibitory effect on the Gly488-Ly353 site, decreasing the Laplacian of the electronic density of the BCP O4-N10. The molecule 2542 showed an inhibitory effect in two regions of interaction of the Tyr491-Glu37 site, acting on the BCPs H30-H33 and O8-H31 while the ligand 2600, in conformation 26, presented a similar effect only on the BCP O8-H31 of that same interactive site. Thus, the data suggest laboratory tests of a combination of molecules that can act at two sites of interaction simultaneously, using the combination of 1629/2542 and 1629/2600 ligands.

Abstract: As the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) pandemic engulfs millions worldwide, the quest for vaccines or drugs against the virus continues. The helicase protein of SARS-CoV-2 represents an attractive target for drug discovery since inhibition of helicase activity can suppress viral replication. Using in silico approaches, we have identified drugs that interact with SARS-CoV-2 helicase based on the presence of amino acid arrangements matching binding sites of drugs in previously annotated protein structures. The drugs exhibiting an RMSD of = 3.0 A were further analyzed using molecular docking, molecular dynamics (MD) simulation, and post-MD analyses. Using these approaches, we found 12 drugs that showed strong interactions with SARS-CoV-2 helicase amino acids. The analyses were performed using the recently available SARS-CoV-2 helicase structure (PDB ID: 5RL6). Based on the MM-GBSA approach, out of the 12 drugs, two drugs, namely posaconazole and grazoprevir, showed the most favorable binding energy, - 54.8 and - 49.1 kcal/mol, respectively. Furthermore, of the amino acids found conserved among all human coronaviruses, 10/11 and 10/12 were targeted by, respectively, grazoprevir and posaconazole. These residues are part of the crucial DEAD-like helicase C and DEXXQc_Upf1-like/ DEAD-like helicase domains. Strong interactions of posaconazole and grazoprevir with conserved amino acids indicate that the drugs can be potent against SARS-CoV-2. Since the amino acids are conserved among the human coronaviruses, the virus is unlikely to develop resistance mutations against these drugs. Since these drugs are already in use, they may be immediately repurposed for SARS-CoV-2 therapy.

Abstract: Identification of the protein targets of hit molecules is essential in the drug discovery process. Target prediction with machine learning algorithms can help accelerate this search, limiting the number of required experiments. However, Drug-Target Interactions databases used for training present high statistical bias, leading to a high number of false positives, thus increasing time and cost of experimental validation campaigns. To minimize the number of false positives among predicted targets, we propose a new scheme for choosing negative examples, so that each protein and each drug appears an equal number of times in positive and negative examples. We artificially reproduce the process of target identification for three specific drugs, and more globally for 200 approved drugs. For the detailed three drug examples, and for the larger set of 200 drugs, training with the proposed scheme for the choice of negative examples improved target prediction results: the average number of false positives among the top ranked predicted targets decreased, and overall, the rank of the true targets was improved.Our method corrects databases' statistical bias and reduces the number of false positive predictions, and therefore the number of useless experiments potentially undertaken.

Abstract: The COVID-19 pandemic has highlighted the need to quickly and reliably prioritize clinically approved compounds for their potential effectiveness for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections. Here, we deployed algorithms relying on artificial intelligence, network diffusion, and network proximity, tasking each of them to rank 6,340 drugs for their expected efficacy against SARS-CoV-2. To test the predictions, we used as ground truth 918 drugs experimentally screened in VeroE6 cells, as well as the list of drugs in clinical trials that capture the medical community's assessment of drugs with potential COVID-19 efficacy. We find that no single predictive algorithm offers consistently reliable outcomes across all datasets and metrics. This outcome prompted us to develop a multimodal technology that fuses the predictions of all algorithms, finding that a consensus among the different predictive methods consistently exceeds the performance of the best individual pipelines. We screened in human cells the top-ranked drugs, obtaining a 62% success rate, in contrast to the 0.8% hit rate of nonguided screenings. Of the six drugs that reduced viral infection, four could be directly repurposed to treat COVID-19, proposing novel treatments for COVID-19. We also found that 76 of the 77 drugs that successfully reduced viral infection do not bind the proteins targeted by SARS-CoV-2, indicating that these network drugs rely on network-based mechanisms that cannot be identified using docking-based strategies. These advances offer a methodological pathway to identify repurposable drugs for future pathogens and neglected diseases underserved by the costs and extended timeline of de novo drug development.

Abstract: In ligand-based drug design, quantitative structure-activity relationship (QSAR) models play an important role in activity prediction. One of the major end points of QSAR models is half-maximal inhibitory concentration (IC50). Experimental IC50 data from various research groups have been accumulated in publicly accessible databases, providing an opportunity for us to use such data in predictive QSAR models. In this study, we focused on using a ranking-oriented QSAR model as a predictive model because relative potency strength within the same assay is solid information that is not based on any mechanical assumptions. We conducted rigorous validation using the ChEMBL database and previously reported data sets. Ranking support vector machine (ranking-SVM) models trained on compounds from similar assays were as good as support vector regression (SVR) with the Tanimoto kernel trained on compounds from all the assays. As effective ways of data integration, for ranking-SVM, integrated compounds should be selected from only similar assays in terms of compounds. For SVR with the Tanimoto kernel, entire compounds from different assays can be incorporated.

Abstract: The main-protease (M(pro)) catalyzes a crucial step for the SARS-CoV-2 life cycle. The recent SARS-CoV-2 presents the main protease (M(CoV2)pro) with 12 mutations compared to SARS-CoV (M(CoV1)pro). Recent studies point out that these subtle differences lead to mobility variances at the active site loops with functional implications. We use metadynamics simulations and a sort of computational analysis to probe the dynamic, pharmacophoric and catalytic environment differences between the monomers of both enzymes. So, we verify how much intrinsic distinctions are preserved in the functional dimer of M(CoV2)pro, as well as its implications for ligand accessibility and optimized drug screening. We find a significantly higher accessibility to open binding conformers in the M(CoV2)pro monomer compared to M(CoV1)pro. A higher hydration propensity for the M(CoV2)pro S2 loop with the A46S substitution seems to exercise a key role. Quantum calculations suggest that the wider conformations for M(CoV2)pro are less catalytically active in the monomer. However, the statistics for contacts involving the N-finger suggest higher maintenance of this activity at the dimer. Docking analyses suggest that the ability to vary the active site width can be important to improve the access of the ligand to the active site in different ways. So, we carry out a multiconformational virtual screening with different ligand bases. The results point to the importance of taking into account the protein conformational multiplicity for new promissors anti M(CoV2)pro ligands. We hope these results will be useful in prospecting, repurposing and/or designing new anti SARS-CoV-2 drugs.Communicated by Ramaswamy H. Sarma.

Abstract: Two thousand nineteen novel coronavirus SARS-CoV-2, the pathogen of COVID-19, has caused a catastrophic pandemic, which has a profound and widespread impact on human lives and social economy globally. However, the molecular perturbations induced by the SARS-CoV-2 infection remain unknown. In this paper, from the perspective of omnigenic, we analyze the properties of the neighborhood perturbed by SARS-CoV-2 in the human interactome and disclose the peripheral and core regions of virus-host network (VHN). We find that the virus-host proteins (VHPs) form a significantly connected VHN, among which highly perturbed proteins aggregate into an observable core region. The non-core region of VHN forms a large scale but relatively low perturbed periphery. We further validate that the periphery is non-negligible and conducive to identifying comorbidities and detecting drug repurposing candidates for COVID-19. We particularly put forward a flower model for COVID-19, SARS and H1N1 based on their peripheral regions, and the flower model shows more correlations between COVID-19 and other two similar diseases in common functional pathways and candidate drugs. Overall, our periphery-core pattern can not only offer insights into interconnectivity of SARS-CoV-2 VHPs but also facilitate the research on therapeutic drugs.

Abstract: BACKGROUND: South African Natural Compounds Database (SANCDB; https://sancdb.rubi.ru.ac.za/ ) is the sole and a fully referenced database of natural chemical compounds of South African biodiversity. It is freely available, and since its inception in 2015, the database has become an important resource to several studies. Its content has been: used as training data for machine learning models; incorporated to larger databases; and utilized in drug discovery studies for hit identifications. DESCRIPTION: Here, we report the updated version of SANCDB. The new version includes 412 additional compounds that have been reported since 2015, giving a total of 1012 compounds in the database. Further, although natural products (NPs) are an important source of unique scaffolds, they have a major drawback due to their complex structure resulting in low synthetic feasibility in the laboratory. With this in mind, SANCDB is, now, updated to provide direct links to commercially available analogs from two major chemical databases namely Mcule and MolPort. To our knowledge, this feature is not available in other NP databases. Additionally, for easier access to information by users, the database and website interface were updated. The compounds are now downloadable in many different chemical formats. CONCLUSIONS: The drug discovery process relies heavily on NPs due to their unique chemical organization. This has inspired the establishment of numerous NP chemical databases. With the emergence of newer chemoinformatic technologies, existing chemical databases require constant updates to facilitate information accessibility and integration by users. Besides increasing the NPs compound content, the updated SANCDB allows users to access the individual compounds (if available) or their analogs from commercial databases seamlessly.

Abstract: MOTIVATION: Drug-target interaction (DTI) prediction is a foundational task for in-silico drug discovery, which is costly and time-consuming due to the need of experimental search over large drug compound space. Recent years have witnessed promising progress for deep learning in DTI predictions. However, the following challenges are still open: (i) existing molecular representation learning approaches ignore the sub-structural nature of DTI, thus produce results that are less accurate and difficult to explain and (ii) existing methods focus on limited labeled data while ignoring the value of massive unlabeled molecular data. RESULTS: We propose a Molecular Interaction Transformer (MolTrans) to address these limitations via: (i) knowledge inspired sub-structural pattern mining algorithm and interaction modeling module for more accurate and interpretable DTI prediction and (ii) an augmented transformer encoder to better extract and capture the semantic relations among sub-structures extracted from massive unlabeled biomedical data. We evaluate MolTrans on real-world data and show it improved DTI prediction performance compared to state-of-the-art baselines. AVAILABILITY AND IMPLEMENTATION: The model scripts are available at https://github.com/kexinhuang12345/moltrans. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.