Current strategies to address data scarcity in artificial intelligence-based drug discovery: A comprehensive review.

Artificial intelligence (AI) has played a vital role in computer-aided drug design (CADD). This development has been further accelerated with the increasing use of machine learning (ML), mainly deep learning (DL), and computing hardware and software advancements. As a result, initial doubts about the application of AI in drug discovery have been dispelled, leading to significant benefits in medicinal chemistry. At the same time, it is crucial to recognize that AI is still in its infancy and faces a few limitations that need to be addressed to harness its full potential in drug discovery. Some notable limitations are insufficient, unlabeled, and non-uniform data, the resemblance of some AI-generated molecules with existing molecules, unavailability of inadequate benchmarks, intellectual property rights (IPRs) related hurdles in data sharing, poor understanding of biology, focus on proxy data and ligands, lack of holistic methods to represent input (molecular structures) to prevent pre-processing of input molecules (feature engineering), etc. The major component in AI infrastructure is input data, as most of the successes of AI-driven efforts to improve drug discovery depend on the quality and quantity of data, used to train and test AI algorithms, besides a few other factors. Additionally, data-gulping DL approaches, without sufficient data, may collapse to live up to their promise. Current literature suggests a few methods, to certain extent, effectively handle low data for better output from the AI models in the context of drug discovery. These are transferring learning (TL), active learning (AL), single or one-shot learning (OSL), multi-task learning (MTL), data augmentation (DA), data synthesis (DS), etc. One different method, which enables sharing of proprietary data on a common platform (without compromising data privacy) to train ML model, is federated learning (FL). In this review, we compare and discuss these methods, their recent applications, and limitations while modeling small molecule data to get the improved output of AI methods in drug discovery. Article also sums up some other novel methods to handle inadequate data.

Multiple risk factors for persistent HBV viraemia in an adult receiving nucleos/tide analogue therapy.

Diagnosing and treating chronic hepatitis B virus (HBV) infection are key interventions to support progress towards elimination of viral hepatitis by 2030. Although nucleos/tide analogue (NA) therapy is typically highly effective, challenges remain for viral load (VL) suppression, including medication access, incomplete adherence and drug resistance. We present a case of a long-term HBV and HIV coinfected adult prescribed with sequential NA therapy regimens, with episodes of breakthrough viraemia. Multiple factors contribute to virological breakthrough, including exposure to old NA agents, initial high HBV VL, therapy interruptions, intercurrent illnesses and potential contribution from resistance mutations. The case underscores the importance of individualised treatment approaches and adherence support in achieving HBV suppression. Furthermore, it emphasises the need for improved clinical pathways addressing education, support and access to care, particularly for marginalised populations. Comprehensive data collection inclusive of under-represented individuals is crucial for maintaining retention in the care cascade and informing effective interventions.

Generative artificial intelligence in drug discovery: basic framework, recent advances, challenges, and opportunities.

There are two main ways to discover or design small drug molecules. The first involves fine-tuning existing molecules or commercially successful drugs through quantitative structure-activity relationships and virtual screening. The second approach involves generating new molecules through de novo drug design or inverse quantitative structure-activity relationship. Both methods aim to get a drug molecule with the best pharmacokinetic and pharmacodynamic profiles. However, bringing a new drug to market is an expensive and time-consuming endeavor, with the average cost being estimated at around $2.5 billion. One of the biggest challenges is screening the vast number of potential drug candidates to find one that is both safe and effective. The development of artificial intelligence in recent years has been phenomenal, ushering in a revolution in many fields. The field of pharmaceutical sciences has also significantly benefited from multiple applications of artificial intelligence, especially drug discovery projects. Artificial intelligence models are finding use in molecular property prediction, molecule generation, virtual screening, synthesis planning, repurposing, among others. Lately, generative artificial intelligence has gained popularity across domains for its ability to generate entirely new data, such as images, sentences, audios, videos, novel chemical molecules, etc. Generative artificial intelligence has also delivered promising results in drug discovery and development. This review article delves into the fundamentals and framework of various generative artificial intelligence models in the context of drug discovery via de novo drug design approach. Various basic and advanced models have been discussed, along with their recent applications. The review also explores recent examples and advances in the generative artificial intelligence approach, as well as the challenges and ongoing efforts to fully harness the potential of generative artificial intelligence in generating novel drug molecules in a faster and more affordable manner. Some clinical-level assets generated form generative artificial intelligence have also been discussed in this review to show the ever-increasing application of artificial intelligence in drug discovery through commercial partnerships.

Dual targeting in prostate cancer with phytoconstituents as a potent lead: a computational approach for novel drug discovery.

Prostate Cancer (PCa) is an abnormal cell growth within the prostate. This condition is the second most widespread malignancy in elderly males and one of the most frequently diagnosed life-threatening conditions. The Androgen receptor signaling pathway played a crucial role in the initiation and spread to increase the risk of PCa. Hence, targeting the AR receptor signaling pathway is a key strategy for a therapeutic plan for PCa. Our study focuses on recognizing potential inhibitors for dual targeting in PCa by using the in-silico approach. In this study, we target the two enzymes that are CYP17A1 (3RUK) and 5α-reductase (3G1R) responsible for PCa, with the help of phytoconstituents. The natural plant contains various phytochemical types produced from secondary metabolites and used as a medical treatment. The in-silico investigation of phytoconstituents and enzymes was done by approaching molecular docking, ADMET analysis, and high-level molecular dynamic simulation used to assess the stability and binding affinities of the protein-ligand complex. Some phytoconstituents, such as Peonidin, Pelargonidin, Malvidin and Berberine show complex has good molecular interaction with protein. The reliability of the docking scores was examined using a molecular dynamic simulation, which revealed that the complex remained stable throughout the simulation, which ranged from 0 to 200 ns. The selected hits may be effective against CYP17A1 (3RUK) and 5α-reductase (3G1R) (PCa) using a computer-aided drug design (CADD) method, which further enables researchers for upcoming in-vivo and in-vitro research, according to our in-silico approach.Communicated by Ramaswamy H. Sarma.

Utility of a buccal swab point-of-care test for the IFNL4 genotype in the era of direct acting antivirals for hepatitis C virus.

BACKGROUND: The CC genotype of the IFNL4 gene is known to be associated with increased Hepatitis C (HCV) cure rates with interferon-based therapy and may contribute to cure with direct acting antivirals. The Genedrive® IFNL4 is a CE marked Point of Care (PoC) molecular diagnostic test, designed for in vitro diagnostic use to provide rapid, real-time detection of IFNL4 genotype status for SNP rs12979860. METHODS: 120 Participants were consented to a substudy comparing IFNL4 genotyping results from a buccal swab analysed on the Genedrive® platform with results generated using the Affymetix UK Biobank array considered to be the gold standard. RESULTS: Buccal swabs were taken from 120 participants for PoC IFNL4 testing and a whole blood sample for genetic sequencing. Whole blood genotyping vs. buccal swab PoC testing identified 40 (33%), 65 (54%), and 15 (13%) had CC, CT and TT IFNL4 genotype respectively. The Buccal swab PoC identified 38 (32%) CC, 64 (53%) CT and 18 (15%) TT IFNL4 genotype respectively. The sensitivity and specificity of the buccal swab test to detect CC vs non-CC was 90% (95% CI 76-97%) and 98% (95% CI 91-100%) respectively. CONCLUSIONS: The buccal swab test was better at correctly identifying non-CC genotypes than CC genotypes. The high specificity of the Genedrive® assay prevents CT/TT genotypes being mistaken for CC, and could avoid patients being identified as potentially 'good responders' to interferon-based therapy.

The impact of pre-existing cross-reactive immunity on SARS-CoV-2 infection and vaccine responses.

Pre-existing cross-reactive immune responses to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) proteins in infection-naive subjects have been described by several studies. In particular, regions of high homology between SARS-CoV-2 and common cold coronaviruses have been highlighted as a likely source of this cross-reactivity. However, the role of such cross-reactive responses in the outcome of SARS-CoV-2 infection and vaccination is currently unclear. Here, we review evidence regarding the impact of pre-existing humoral and T cell immune responses to outcomes of SARS-CoV-2 infection and vaccination. Furthermore, we discuss the importance of conserved coronavirus epitopes for the rational design of pan-coronavirus vaccines and consider cross-reactivity of immune responses to ancestral SARS-CoV-2 and SARS-CoV-2 variants, as well as their impact on COVID-19 vaccination.

Strategic treatment optimization for HCV (STOPHCV1): a randomised controlled trial of ultrashort duration therapy for chronic hepatitis C.

Background: The world health organization (WHO) has identified the need for a better understanding of which patients with hepatitis C virus (HCV) can be cured with ultrashort course HCV therapy. Methods: A total of 202 individuals with chronic HCV were randomised to fixed-duration shortened therapy (8 weeks) vs variable-duration ultrashort strategies (VUS1/2). Participants not cured following first-line treatment were retreated with 12 weeks' sofosbuvir/ledipasvir/ribavirin. The primary outcome was sustained virological response 12 weeks (SVR12) after first-line treatment and retreatment. Participants were factorially randomised to receive ribavirin with first-line treatment. Results: All evaluable participants achieved SVR12 overall (197/197, 100% [95% CI 98-100]) demonstrating non-inferiority between fixed-duration and variable-duration strategies (difference 0% [95% CI -3.8%, +3.7%], 4% pre-specified non-inferiority margin). First-line SVR12 was 91% [86%-97%] (92/101) for fixed-duration vs 48% [39%-57%] (47/98) for variable-duration, but was significantly higher for VUS2 (72% [56%-87%] (23/32)) than VUS1 (36% [25%-48%] (24/66)). Overall, first-line SVR12 was 72% [65%-78%] (70/101) without ribavirin and 68% [61%-76%] (69/98) with ribavirin (p=0.48). At treatment failure, the emergence of viral resistance was lower with ribavirin (12% [2%-30%] (3/26)) than without (38% [21%-58%] (11/29), p=0.01). Conclusions: Unsuccessful first-line short-course therapy did not compromise retreatment with sofosbuvir/ledipasvir/ribavirin (100% SVR12). SVR12 rates were significantly increased when ultrashort treatment varied between 4-7 weeks rather than 4-6 weeks. Ribavirin significantly reduced resistance emergence in those failing first-line therapy. ISRCTN Registration: 37915093 (11/04/2016).

Novel, selective acrylamide linked quinazolines for the treatment of double mutant EGFR-L858R/T790M Non-Small-Cell lung cancer (NSCLC).

T790M mutation is the most common mechanism of acquired resistance to first-generation epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs). To overcome this resistance, 4-anilinoquinazoline-based irreversible inhibitors afatinib, dacomitinib has been developed. However, the clinical application of these irreversible inhibitors is limited due to its narrow selectivity against L858R/T790M mutant EGFR. In an attempt to develop potent and selective EGFR T790M inhibitors, we have designed and synthesized two series of novel acrylamide linked quinazolines. Among them, compounds 2i (IC50 0.171 µM) and 11h (IC50 0.159 µM) were identified as potent compounds, which displayed selective and potent anti-proliferative activity on gefitinib-resistant cell line NCI-H1975 as compared to the gefitinib and WZ4002 in cellular assay. Furthermore, a molecular dynamic simulation of 11h was carried out to assess the stability to form a complex with the L858R/T790M EGFR Kinase domain, which demonstrated that complex was stable for the 100 ns and form strong crucial covalent binding contacts with the thiol group of Cys797 residue. Finally, satisfactory in silico pharmacokinetics properties of 2i, 11h and 11i compounds were predicted. The synthesized compounds were also evaluated for in vitro cytotoxic activity/hepatotoxicity against HepG2 cell line through MTT assay. The results revealed that compounds exhibited lower cytotoxicity to HepG2 cells.

Broad and strong memory CD4 + and CD8 + T cells induced by SARS-CoV-2 in UK convalescent COVID-19 patients.

COVID-19 is an ongoing global crisis in which the development of effective vaccines and therapeutics will depend critically on understanding the natural immunity to the virus, including the role of SARS-CoV-2-specific T cells. We have conducted a study of 42 patients following recovery from COVID-19, including 28 mild and 14 severe cases, comparing their T cell responses to those of 16 control donors. We assessed the immune memory of T cell responses using IFNγ based assays with overlapping peptides spanning SARS-CoV-2 apart from ORF1. We found the breadth, magnitude and frequency of memory T cell responses from COVID-19 were significantly higher in severe compared to mild COVID-19 cases, and this effect was most marked in response to spike, membrane, and ORF3a proteins. Total and spike-specific T cell responses correlated with the anti-Spike, anti-Receptor Binding Domain (RBD) as well as anti-Nucleoprotein (NP) endpoint antibody titre (p<0.001, <0.001 and =0.002). We identified 39 separate peptides containing CD4 + and/or CD8 + epitopes, which strikingly included six immunodominant epitope clusters targeted by T cells in many donors, including 3 clusters in spike (recognised by 29%, 24%, 18% donors), two in the membrane protein (M, 32%, 47%) and one in the nucleoprotein (Np, 35%). CD8+ responses were further defined for their HLA restriction, including B*4001-restricted T cells showing central memory and effector memory phenotype. In mild cases, higher frequencies of multi-cytokine producing M- and NP-specific CD8 + T cells than spike-specific CD8 + T cells were observed. They furthermore showed a higher ratio of SARS-CoV-2-specific CD8 + to CD4 + T cell responses. Immunodominant epitope clusters and peptides containing T cell epitopes identified in this study will provide critical tools to study the role of virus-specific T cells in control and resolution of SARS-CoV-2 infections. The identification of T cell specificity and functionality associated with milder disease, highlights the potential importance of including non-spike proteins within future COVID-19 vaccine design.

Amino Acid Substitutions in Genotype 3a Hepatitis C Virus Polymerase Protein Affect Responses to Sofosbuvir.

BACKGROUND & AIMS: Sofosbuvir is a frequently used pan-genotype inhibitor of hepatitis C virus (HCV) polymerase. This drug eliminates most chronic HCV infections, and resistance-associated substitutions in the polymerase are rare. However, HCV genotype 3 responds slightly less well to sofosbuvir-based therapies than other genotypes. We collected data from England's National Health Service Early Access Program to search for virus factors associated with sofosbuvir treatment failure. METHODS: We collected patient serum samples and used the capture-fusion assay to assess viral sensitivity to sofosbuvir in 14 HCV genotype 3 samples. We identified polymorphisms associated with reduced response and created modified forms of HCV and replicons containing the substitutions of interest and tested their sensitivity to sofosbuvir and ribavirin. We examined the effects of these polymorphisms by performing logistic regression multivariate analysis on their association with sustained virologic response in a separate cohort of 411 patients with chronic HCV genotype 3 infection who had been treated with sofosbuvir and ribavirin, with or without pegylated interferon. RESULTS: We identified a substitution in the HCV genotype 3a NS5b polymerase at amino acid 150 (alanine [A] to valine [V]), V at position 150 was observed in 42% of patients) with a reduced response to sofosbuvir in virus replication assays. In patients treated with sofosbuvir-containing regimens, the A150V variant was associated with a reduced response to treatment with sofosbuvir and ribavirin, with or without pegylated interferon. In 326 patients with V at position 150, 71% achieved an sustained virologic response compared to 88% with A at position 150. In cells, V at position 150 reduced the response to sofosbuvir 7-fold. We found that another rare substitution, glutamic acid (E) at position 206, significantly reduced the response to sofosbuvir (8.34-fold reduction); the combinations of V at position 150 and E at position 206 reduced the virus response to sofosbuvir 35.77-fold. Additionally, in a single patient, we identified 5 rare polymorphisms that reduced sensitivity to sofosbuvir our cell system. CONCLUSIONS: A common polymorphism, V at position 150 in the HCV genotype 3a NS5b polymerase, combined with other variants, reduces the virus response to sofosbuvir. Clinically, infection with HCV genotype 3 containing this variant reduces odds of sustained virologic response. In addition, we identified rare combinations of variants in HCV genotype 3 that reduce response to sofosbuvir.

Design and synthesis of novel 2,4-disubstituted aminopyrimidines: reversible non-covalent T790M EGFR inhibitors.

Cardiotoxicity is one amongst the adverse effect of Osimertinib delineate in clinical trials and related to escalating doses. To triumph over the drawbacks of Osimertinib, in this study, we tend to delineate the design, synthesis, in vitro biological analysis of a series of novel reversible selective T790M inhibitors with minimal cardiotoxicity. Amongst the virtually sorted compounds; compound 18 and 74 have been located to be the foremost active compounds of the series with IC50 value of 0.88, 0.92 µM in cellular assay and 0.56, 0.62 µM in enzymatic assay, against double mutant L858R/T790M EGFR. Additionally, they showed much less affinity toward wild-type (WT)-EGFR with minimal cardiotoxicity.

Recent updates on third generation EGFR inhibitors and emergence of fourth generation EGFR inhibitors to combat C797S resistance.

EGFR T790M mutation leads to resistance to most of clinically available EGFR TKIs. Third-generation EGFR TKIs against the T790M mutation have been in active clinical development, which includes osimertinib, rociletinib, HM61713, ASP8273, EGF816, and PF-06747775. On the other hand recently EGFR C797S mutation was reported to be a leading mechanism of resistance to the third-generation inhibitors. The C797S mutation appears to be an ideal target for overcoming the acquired resistance to the third generation inhibitors. This review summarizes the third generation inhibitors, synthesis, their mechanism of resistance and latest development on the discovery of a fourth-generation EGFR TKIs and U to Y allosteric strategies to combat the C797S EGFR resistance problem.

Design and synthesis of quinazolinones as EGFR inhibitors to overcome EGFR resistance obstacle.

The epidermal growth factor receptor (EGFR) T790M mutant is found in about 50% of clinically acquired resistance to gefitinib among patients with non-small cell lung cancer (NSCLC). New derivatives of 4(3H)-quinazolinones were synthesized and evaluated for their inhibitory activity against NSCLC. The results of the study demonstrated that compound 79, 7-chloro-3-(5-(4-methoxyphenyl)-1,3,4-thiadiazol-2-yl)-2-phenylquinazolin-4(3H)-one was found to be the most potent compounds of the series with IC50 value of 0.031µM against mutant T790M/L858R EGFR. Compounds 15, 51, 73, 75, 78, 79 and 96 were less potent against A549 (WT EGFR and k-Ras mutation) and HT-29 (non-special gene type) cells, showing a high safety index. The obtained results showed that compounds 15, 51, 73, 75, 78, 79 and 96 could be the promising template to overcome drug resistance mediated by the EGFR T790 Mutant.