Abstract
Background
A Virtual Screening algorithm has to adapt to the different stages of this process. Early screening needs to ensure that all bioactive compounds are ranked in the first positions ...despite of the number of false positives, while a second screening round is aimed at increasing the prediction accuracy.
Results
A novel CNN architecture is presented to this aim, which predicts bioactivity of candidate compounds on CDK1 using a combination of molecular fingerprints as their vector representation, and has been trained suitably to achieve good results as regards both enrichment factor and accuracy in different screening modes (98
.
55% accuracy in active-only selection, and 98
.
88% in high precision discrimination).
Conclusion
The proposed architecture outperforms state-of-the-art ML approaches, and some interesting insights on molecular fingerprints are devised.
In the last year, the COVID-19 pandemic has highly affected the lifestyle of the world population, encouraging the scientific community towards a great effort on studying the infection molecular ...mechanisms. Several vaccine formulations are nowadays available and helping to reach immunity. Nevertheless, there is a growing interest towards the development of novel anti-covid drugs. In this scenario, the main protease (Mpro) represents an appealing target, being the enzyme responsible for the cleavage of polypeptides during the viral genome transcription. With the aim of sharing new insights for the design of novel Mpro inhibitors, our research group developed a machine learning approach using the support vector machine (SVM) classification. Starting from a dataset of two million commercially available compounds, the model was able to classify two hundred novel chemo-types as potentially active against the viral protease. The compounds labelled as actives by SVM were next evaluated through consensus docking studies on two PDB structures and their binding mode was compared to well-known protease inhibitors. The best five compounds selected by consensus docking were then submitted to molecular dynamics to deepen binding interactions stability. Of note, the compounds selected via SVM retrieved all the most important interactions known in the literature.
The family of protein kinases comprises more than 500 genes involved in numerous functions. Hence, their physiological dysfunction has paved the way toward drug discovery for cancer, cardiovascular, ...and inflammatory diseases. As a matter of fact, Kinase binding sites high similarity has a double role. On the one hand it is a critical issue for selectivity, on the other hand, according to poly-pharmacology, a synergistic controlled effect on more than one target could be of great pharmacological interest. Another important aspect of binding similarity is the possibility of exploit it for repositioning of drugs on targets of the same family. In this study, we propose our approach called Kinase drUgs mAchine Learning frAmework (KUALA) to automatically identify kinase active ligands by using specific sets of molecular descriptors and provide a multi-target priority score and a repurposing threshold to suggest the best repurposable and non-repurposable molecules. The comprehensive list of all kinase-ligand pairs and their scores can be found at https://github.com/molinfrimed/multi-kinases .
Glypican-3 (GPC-3) is a heparin sulfate proteoglycan located extracellularly and anchored to the cell membrane of transformed hepatocytes. GPC-3 is not expressed in normal or cirrhotic liver tissue ...but is overexpressed in hepatocellular carcinoma (HCC). Because of this, GPC-3 is one of the most important emerging immunotargets for treatment and as an early detection marker of HCC. To determine if GPC-3 domains associated with serum small extracellular vesicles (sEVs) could be used as an HCC diagnostic marker, we predicted in silico GPC-3 structural properties and tested for the presence of its full-length form and/or cleaved domains in serum sEVs isolated from patients with HCC. Structural analysis revealed that the Furin cleavage site of GPC-3 is exposed and readily accessible, suggesting the facilitation of GPC-3 cleavage events. Upon isolation of sEVs from both hepatocytes, culture media and serum of patients with HCC were studied for GPC-3 content. This data suggests that Furin-dependent GPC-3 cleaved domains could be a powerful tool for detection of initial stages of HCC and serve as a predictor for disease prognosis.
In recent years, the debate in the field of applications of Deep Learning to Virtual Screening has focused on the use of neural embeddings with respect to classical descriptors in order to encode ...both structural and physical properties of ligands and/or targets. The attention on embeddings with the increasing use of Graph Neural Networks aimed at overcoming molecular fingerprints that are short range embeddings for atomic neighborhoods. Here, we present EMBER, a novel molecular embedding made by seven molecular fingerprints arranged as different "spectra" to describe the same molecule, and we prove its effectiveness by using deep convolutional architecture that assesses ligands' bioactivity on a data set containing twenty protein kinases with similar binding sites to CDK1. The data set itself is presented, and the architecture is explained in detail along with its training procedure. We report experimental results and an explainability analysis to assess the contribution of each fingerprint to different targets.
A new series of imidazo2,1-b1,3,4thiadiazole derivatives was efficiently synthesized and screened for their in vitro antiproliferative activity on a panel of pancreatic ductal adenocarcinoma (PDAC) ...cells, including SUIT-2, Capan-1 and Panc-1. Compounds 9c and 9l, showed relevant in vitro antiproliferative activity on all three pre-clinical models with half maximal inhibitory concentration (IC50) ranging from 5.11 to 10.8 µM, while the compounds 9e and 9n were active in at least one cell line. In addition, compound 9c significantly inhibited the migration rate of SUIT-2 and Capan-1 cells in the scratch wound-healing assay. In conclusion, our results will support further studies to increase the library of imidazo 2,1-b1,3,4 thiadiazole derivatives for deeper understanding of the relationship between biological activity of the compounds and their structures in the development of new antitumor compounds against pancreatic diseases.
We present a new approach that incorporates flexibility based on extensive MD simulations of protein–ligand complexes into structure-based pharmacophore modeling and virtual screening. The approach ...uses the multiple coordinate sets saved during the MD simulations and generates for each frame a pharmacophore model. Pharmacophore models with the same pharmacophore features are pooled. In this way the high number of pharmacophore models that results from the MD simulation is reduced to only a few hundred representative pharmacophore models. Virtual screening runs are performed with every representative pharmacophore model; the screening results are combined and rescored to generate a single hit-list. The score for a particular molecule is calculated based on the number of representative pharmacophore models which classified it as active. Hence, the method is called common hits approach (CHA). The steps between the MD simulation and the final hit-list are performed automatically and without user interaction. We test the performance of CHA for virtual screening using screening databases with active and inactive compounds for 40 protein–ligand systems. The results of the CHA are compared to the (i) median screening performance of all representative pharmacophore models of protein–ligand systems, as well as to the virtual screening performance of (ii) a random classifier, (iii) the pharmacophore model derived from the experimental structure in the PDB, and (iv) the representative pharmacophore model appearing most frequently during the MD simulation. For the 34 (out of 40) protein–ligand complexes, for which at least one of the approaches was able to perform better than a random classifier, the highest enrichment was achieved using CHA in 68% of the cases, compared to 12% for the PDB pharmacophore model and 20% for the representative pharmacophore model appearing most frequently. The availabilithy of diverse sets of different pharmacophore models is utilized to analyze some additional questions of interest in 3D pharmacophore-based virtual screening.
Molecular dynamics simulations of twelve protein—ligand systems were used to derive a single, structure based pharmacophore model for each system. These merged models combine the information from the ...initial experimental structure and from all snapshots saved during the simulation. We compared the merged pharmacophore models with the corresponding PDB pharmacophore models, i.e., the static models generated from an experimental structure in the usual manner. The frequency of individual features, of feature types and the occurrence of features not present in the static model derived from the experimental structure were analyzed. We observed both pharmacophore features not visible in the traditional approach, as well as features which disappeared rapidly during the molecular dynamics simulations and which may well be artifacts of the initial PDB structure-derived pharmacophore model. Our approach helps mitigate the sensitivity of structure based pharmacophore models to the single set of coordinates present in the experimental structure. Further, the frequency with which specific features occur during the MD simulation may aid in ranking the importance of individual features.
Display omitted
•Pharmacophore feature types display varying stability in MD simulations.•Features obtained from the crystal structure are on average stable.•Some features obtained from the crystal structure appear less than 10%.•Frequency information from MD simulations can be used to add/remove features.
Have you a compound in your lab, which was not successful against the designed target, or a drug that is no more attractive? The drug repurposing represents the right way to reconsider them. It can ...be defined as the modern and rationale approach of the traditional methods adopted in drug discovery, based on the knowledge, insight and luck, alias known as serendipity. This repurposing approach can be applied both in silico and in wet. In this review we report the molecular modeling facilities that can be of huge support in the repurposing of drugs and/or unsuccessful lead compounds. In the last decades, different methods were proposed to help the scientists in drug design and in drug repurposing. The steps strongly depend on the approach applied. It could be a ligand or a structure based method, correlated to the use of specific means. These processes, starting from a compound with potential therapeutic properties and a sizeable number of toxicity passed tests, can successfully speed up the very slow development of a molecule from bench to market. Herein, we discuss the facilities available to date, classifying them by methods and types. We have reported a series of databases, ligand and structure stand-alone software, and of web-based tools, which are free accessible to scientific community. This review does not claim to be exhaustive, but can be of interest to help in drug repurposing through in silico methods, as a valuable tool for the medicinal chemistry community.
Immunoproteasome inhibition is a promising strategy for the treatment of hematological malignancies, autoimmune diseases, and inflammatory diseases. The design of non-covalent inhibitors of the ...immunoproteasome β1i/β5i catalytic subunits could be a novel approach to avoid the drawbacks of the known covalent inhibitors, such as toxicity due to off-target binding. In this work, we report the biological evaluation of thirty-four compounds selected from a commercially available collection. These hit compounds are the outcomes of a virtual screening strategy including a dynamic pharmacophore modeling approach onto the β1i subunit and a pharmacophore/docking approach onto the β5i subunit. The computational studies were first followed by in vitro enzymatic assays at 100 μM. Only compounds capable of inhibiting the enzymatic activity by more than 50% were characterized in detail using Tian continuous assays, determining the dissociation constant (
) of the non-covalent complex where
is also the measure of the binding affinity. Seven out of thirty-four hits showed to inhibit β1i and/or β5i subunit. Compound
is the most active on the β1i subunit with
= 11.84 ± 1.63 µM, and compound
showed
= 12.50 ± 0.77 µM on the β5i subunit. Compound
showed inhibitory activity on both subunits (
= 12.53 ± 0.18 and
= 31.95 ± 0.81 on the β1i subunit and β5i subunit, respectively). The induced fit docking analysis revealed interactions with Thr1 and Phe31 of β1i subunit and that represent new key residues as reported in our previous work. Onto β5i subunit, it interacts with the key residues Thr1, Thr21, and Tyr169. This last hit compound identified represents an interesting starting point for further optimization of β1i/β5i dual inhibitors of the immunoproteasome.