Prediction of KRASG12C inhibitors using conjoint fingerprint and machine learning-based QSAR models

Evolution of computational techniques against various KRAS mutants in search for therapeutic drugs: a review article

KRAS was (Kirsten rat sarcoma viral oncogene homolog) revealed as an important target in current therapeutic cancer research because alteration of RAS (rat sarcoma viral oncogene homolog) protein has a critical role in malignant modification, tumor angiogenesis, and metastasis. For cancer treatment, designing competitive inhibitors for this attractive target was difficult. Nevertheless, computational investigations of the protein's dynamic behavior displayed the existence of temporary pockets that could be used to design allosteric inhibitors. The last decade witnessed intensive efforts to discover KRAS inhibitors. In 2021, the first KRAS G12C covalent inhibitor, AMG 510, received FDA (Food and drug administration) approval as an anticancer medication that paved the path for future treatment strategies against this target. Computer-aided drug designing discovery has long been used in drug development research targeting different KRAS mutants. In this review, the major breakthroughs in computational methods adapted to discover novel compounds for different mutations have been discussed. Undoubtedly, virtual screening and molecular dynamic (MD) simulation and molecular docking are the most considered approach, producing hits that can be employed in subsequent refinements. After comprehensive analysis, Afatinib and Quercetin were computationally identified as hits in different publications. Several authors conducted covalent docking studies with acryl amide warheads groups containing inhibitors. Future studies are needed to demonstrate their true potential. In-depth studies focusing on various allosteric pockets demonstrate that the switch I/II pocket is a suitable site for drug designing. In addition, machine learning and deep learning based approaches provide new insights for developing anti-KRAS drugs. We believe that this review provides extensive information to researchers globally and encourages further development in this particular area of research.

In Silico Identification of Putative Allosteric Pockets and Inhibitors for the KRASG13D-SOS1 Complex in Cancer Therapy

RAS mutations occur in about 30% of human cancers, leading to enhanced RAS signaling and tumor growth. KRAS is the most commonly mutated oncogene in human tumors, especially lung, pancreatic, and colorectal cancers. Direct targeting of KRAS is difficult due to its highly conserved sequence; but, its complex with the guanine nucleotide exchange factor Son of Sevenless (SOS) 1 promises an attractive target for inhibiting RAS-mediated signaling. Here, we first revealed putative allosteric binding sites of the SOS1, KRASG12C-SOS1 complex, and the ternary KRASG13D-SOS1 complex structures using two network-based models, the essential site scanning analysis and the residue interaction network model. The results enabled us to identify two new putative allosteric pockets for the ternary KRASG13D-SOS1 complex. These were then screened together with the known ligand binding site against the natural compounds in the InterBioScreen (IBS) database using the Glide software package developed by Schrödinger, Inc. The docking poses of seven hit compounds were assessed using 400 ns long molecular dynamics (MD) simulations with two independent replicas using Desmond, coupled with thermal MM-GBSA calculations for the estimation of the binding free energy values. The structural skeleton of the seven proposed compounds consists of different functional groups and heterocyclic rings that possess anti-cancer activity and exhibit persistent interactions with key residues in binding pockets throughout the MD simulations. STOCK1N-09823 was determined as the most promising hit that promoted the disruption of the interactions R73 (chain A)/N879 and R73 (chain A)/Y884, which are key for SOS1-mediated KRAS activation.

Machine-learning-assisted high-throughput identification of potent and stable neutralizing antibodies against all four dengue virus serotypes

Several computational methods have been developed to identify neutralizing antibodies (NAbs) covering four dengue virus serotypes (DENV-1 to DENV-4); however, limitations of the dataset and the resulting performance remain. Here, we developed a new computational framework to predict potent and stable NAbs against DENV-1 to DENV-4 using only antibody (CDR-H3) and epitope sequences as input. Specifically, our proposed computational framework employed sequence-based ML and molecular dynamic simulation (MD) methods to achieve more accurate identification. First, we built a novel dataset (n = 1108) by compiling the interactions of CDR-H3 and epitope sequences with the half maximum inhibitory concentration (IC50) values, which represent neutralizing activities. Second, we achieved an accurately predictive ML model that showed high AUC values of 0.879 and 0.885 by tenfold cross-validation and independent tests, respectively. Finally, our computational framework could be applied to filter approximately 2.5 million unseen antibodies into two final candidates that showed strong and stable binding to all four serotypes. In addition, the most potent and stable candidate (1B3B9_V21) was evaluated for its development potential as a therapeutic agent by molecular docking and MD simulations. This study provides an antibody computational approach to facilitate the high-throughput identification of NAbs and accelerate the development of therapeutic antibodies.

Insights into the structure-activity relationship of pyrimidine-sulfonamide analogues for targeting BRAF V600E protein

B-rapidly accelerated fibrosarcoma (BRAF) V600E plays a crucial role in the progression of cutaneous melanoma. Core structures of BRAF V600E inhibitors are based on pyrimidine-sulfonamide scaffolds. Exploring the QSAR of these structures can improve our understanding of BRAF V600E inhibitor drug design. This study utilized machine learning-based QSAR to elucidate chemical substructures of pyrimidine-sulfonamide analogues that correlated to the BRAF V600E inhibitory activity. The findings indicate that the support vector regression (SVR) combined with 15 fingerprints achieved the highest statistical performances in terms of goodness-of-fit, robustness, and predictability. Nine key fingerprints from pyrimidine-sulfonamide analogues were identified to exert the BRAF V600E inhibitory activity. These key fingerprints were validated using network-based activity cliff landscape and molecular docking. Together, the developed algorithm can serve as a screening tool for designing BRAF V600E inhibitors. To further utilize this model, we deployed our developed algorithm at https://qsarlabs.com/#braf.

Ensemble Quantitative Read-Across Structure-Activity Relationship Algorithm for Predicting Skin Cytotoxicity

Read-across (RA) and quantitative structure-activity relationship (QSAR) are two alternative methods commonly used to fill data gaps in chemical registrations. These approaches use physicochemical properties or molecular fingerprints of source substances to predict the properties of unknown substances that have similar chemical structures or physicochemical properties. Research on RA and QSAR is essential to minimize the time, money, and animal testing needed to determine biological properties that are not currently known. This study developed a stacked ensemble quantitative read-across structure-activity relationship algorithm (enQRASAR) for predicting skin irritation toxicity based on negative log cell viability inhibition concentration at 50% (pIC50) against skin keratinocytes as the end point. The goodness-of-fit and predictability of this algorithm were validated using leave-one-out cross-validation and external test data sets. The results obtained were statistically reliable in terms of goodness-of-fit, robustness, and predictability metrics. Additionally, the developed model demonstrated a low prediction error when predicting FDA-approved drugs. These results confirm that the enQRASAR algorithm can be used to predict skin cytotoxicity of chemicals. Therefore, this model was publicly available to further facilitate toxicity predictions of unknown compounds in chemical registrations.

Stacked ensemble learning on HaCaT cytotoxicity for skin irritation prediction: A case study on dipterocarpol

Skin irritation is an adverse effect associated with various substances, including chemicals, drugs, or natural products. Dipterocarpol, extracted from Dipterocarpus alatus, contains several skin benefits notably anticancer, wound healing, and antibacterial properties. However, the skin irritation of dipterocarpol remains unassessed. Quantitative structure-activity relationship (QSAR) is a recommended tool for toxicity assessment involving less time, money, and animal testing to access unavailable acute toxicity data. Therefore, our study aimed to develop a highly accurate machine learning-based QSAR model for predicting skin irritation. We utilized a stacked ensemble learning model with 1064 chemicals. We also adhered to the recommendations from the OECD for QSAR validation. Subsequently, we used the proposed model to explore the cytotoxicity of dipterocarpol on keratinocytes. Our findings indicate that the model displayed promising statistical quality in terms of accuracy, precision, and recall in both 10-fold cross-validation and test datasets. Moreover, the model predicted that dipterocarpol does not have skin irritation, which was confirmed by the cell-based assay. In conclusion, our proposed model can be applied for the risk assessment of skin irritation in untested compounds that fall within its applicability domain. The web application of this model is available at https://qsarlabs.com/#stackhacat.