Discovery Methods Research Articles

Application of artificial intelligence and machine learning methods in drug discovery and development

Application of artificial intelligence and machine learning methods in drug discovery and development

Raman spectroscopic imaging – practical applications in industrial pharmacy and medical diagnostics: a review.

Currently, Raman imaging as a non-destructive technique providing knowledge on structural characterization of the investigated samples is gaining more and more interest among industrial formulation and process scientists. In the pharmaceutical sector, the creation of novel products and formulations necessitates flexible and adaptable analysis, and increasingly also imaging capabilities. In this paper, the attention is directed toward the research progress, that has occurred in the application of non-invasive Raman chemical imaging techniques in industrial pharmacy and medical diagnostics in recent years. This work aims to explore the potential of Raman imaging as a knowledge discovery method for the non-destructive evaluation of pharmaceutical formulations as well as human cells to support medical diagnostics in clinical practice. A review of the latest scientific papers indicates that the applications of Raman imaging in pharmaceutical sciences and medical diagnostics are diverse and far-reaching from ensuring the quality of pharmaceutical formulations to studying complex drug delivery systems, monitoring manufacturing processes, and medical diagnostics of cancers and other diseases. As technology continues to evolve, the integration of Raman imaging into pharmaceutical and clinical practices is expected to expand, providing new insights and solutions to the challenges faced by the pharmaceutical industry and medical diagnostics.

A Review: Drug Discovery Methods Based on Artificial Intelligence

A Review: Drug Discovery Methods Based on Artificial Intelligence

TumFlow: An AI Model for Predicting New Anticancer Molecules.

Melanoma is the fifth most common cancer in the United States. Conventional drug discovery methods are inherently time-consuming and costly, which imposes significant limitations. However, the advent of Artificial Intelligence (AI) has opened up new possibilities for simulating and evaluating numerous drug candidates, thereby mitigating the requisite time and resources. In this context, normalizing flow models by employing machine learning techniques to create new molecular structures holds promise for accelerating the discovery of effective anticancer therapies. This manuscript introduces TumFlow, a novel AI model designed to generate new molecular entities with potential therapeutic value in cancer treatment. It has been trained on the NCI-60 dataset, encompassing thousands of molecules tested across 60 tumour cell lines, with an emphasis on the melanoma SK-MEL-28 cell line. The model successfully generated new molecules with predicted improved efficacy in inhibiting tumour growth while being synthetically feasible. This represents a significant advancement over conventional generative models, which often produce molecules that are challenging or impossible to synthesize. Furthermore, TumFlow has also been utilized to optimize molecules known for their efficacy in clinical melanoma treatments. This led to the creation of novel molecules with a predicted enhanced likelihood of effectiveness against melanoma, currently undocumented on PubChem.

Bayesian network structure learning with a new ensemble weights and edge constraints setting mechanism

Bayesian networks (BNs) are highly effective in handling uncertain problems, which can assist in decision-making by reasoning with limited and incomplete information. Learning a faithful directed acyclic graph (DAG) from a large number of complex samples of a joint distribution is currently a challenging combinatorial problem. Due to the growing volume and complexity of data, some Bayesian structure learning algorithms are ineffective and lack the necessary precision to meet the required needs. In this paper, we propose a new PCCL-CC algorithm. To ensure the accuracy of the network structure, we introduce the new ensemble weights and edge constraints setting mechanism. In this mechanism, we employ a method that estimates the interaction between network nodes from multiple perspectives and divides the learning process into multiple stages. We utilize an asymmetric weighted ensemble method and adaptively adjust the network structure. Additionally, we propose a causal discovery method that effectively utilizes the causal relationships among data samples to correct the network structure and mitigate the influence of Markov equivalence classes (MEC). Experimental results on real datasets demonstrate that our approach outperforms state-of-the-art methods.

Discovering urinary protein biomarkers for hepatocellular carcinoma in the population of patients with chronic hepatitis B: A single-center case-control study.

e16230 Background: Chronic hepatitis B (CHB) can result in hepatocellular carcinoma (HCC) imposing a substantial health and economic burden worldwide. Early detection of hepatitis B virus-related HCC (HBV-HCC) in CHB with potential biomarkers has emerged as an urgent and challenging task. Recent progression of urinary proteomics provides a promising method for HBV-HCC biomarker discovery. Methods: This single-center case-control study, conducted at Peking Union Medical College Hospital, enrolled 33 patients with HBV-HCC, 26 patients with CHB, 30 healthy controls, and 33 controls with other types of cancer. Urine proteomes were analyzed using liquid chromatography with tandem mass spectrometry by data-dependent acquisition or parallel reaction monitoring, respectively. Differential analysis was carried out with limma package in R software, and the origin of selected proteins was confirmed with immunohistochemistry of surgical specimens. The diagnostic performance of the biomarker panel was evaluated with area under receiver operator curves (AUROC). Results: 21 urinary proteins were significantly upregulated in HBV-HCC compared with CHB samples, playing key roles in vesicular transportation, cytoskeleton-extracellular matrix interaction, and tumor development according to enrichment analysis. Immunohistochemistry validated the differential expression of 4 urinary proteins (4UP: HNRNPM, ADGRG1, APOC1, and RALA) across cancerous and adjacent normal liver tissues. These proteins were specifically found in HBV-HCC but not in CHB, healthy control, or other cancer controls. The AUROCs of 4UP individually (0.642 for APOC1, 0.780 for HNRNPM, 0.755 for ADGRG1, 0.729 for RALA) were no worse than that of serum α-fetoprotein (AUROC = 0.751), and the 4UP panel unified via logistic regression showed an AUROC of 0.828. Conclusions: Urinary proteomics revealed HNRNPM, ADGRG1, APOC1, and RALA as biomarkers to distinguish HBV-HCC from CHB. The 4UP could be traced back to HBV-HCC tumor cells, shedding light on pathogenesis of HBV-HCC. [Table: see text]

Artificial Intelligence in Drug Identification and Validation: A Scoping Review.

The end-to-end process in the discovery of drugs involves therapeutic candidate identification, validation of identified targets, identification of hit compound series, lead identification and optimization, characterization, and formulation and development. The process is lengthy, expensive, tedious, and inefficient, with a large attrition rate for novel drug discovery. Today, the pharmaceutical industry is focused on improving the drug discovery process. Finding and selecting acceptable drug candidates effectively can significantly impact the price and profitability of new medications. Aside from the cost, there is a need to reduce the end-to-end process time, limiting the number of experiments at various stages. To achieve this, artificial intelligence (AI) has been utilized at various stages of drug discovery. The present study aims to identify the recent work that has developed AI-based models at various stages of drug discovery, identify the stages that need more concern, present the taxonomy of AI methods in drug discovery, and provide research opportunities. From January 2016 to September 1, 2023, the study identified all publications that were cited in the electronic databases including Scopus, NCBI PubMed, MEDLINE, Anthropology Plus, Embase, APA PsycInfo, SOCIndex, and CINAHL. Utilising a standardized form, data were extracted, and presented possible research prospects based on the analysis of the extracted data.

Accelerating Molecular Docking using Machine Learning Methods.

Virtual screening (VS) is one of the well-established approaches in drug discovery which speeds up the search for a bioactive molecule and, reduces costs and efforts associated with experiments. VS helps to narrow down the search space of chemical space and allows selecting fewer and more probable candidate compounds for experimental testing. Docking calculations are one of the commonly used and highly appreciated structure-based drug discovery methods. Databases for chemical structures of small molecules have been growing rapidly. However, at the moment virtual screening of large libraries via docking is not very common. In this work, we aim to accelerate docking studies by predicting docking scores without explicitly performing docking calculations. We experimented with an attention based long short-term memory (LSTM) neural network for an efficient prediction of docking scores as well as other machine learning models such as XGBoost. By using docking scores of a small number of ligands we trained our models and predicted docking scores of a few million molecules. Specifically, we tested our approaches on 11 datasets that were produced from in-house drug discovery studies. On average, by training models using only 7000 molecules we predicted docking scores of approximately 3.8 million molecules with R2 (coefficient of determination) of 0.77 and Spearman rank correlation coefficient of 0.85. We designed the system with ease of use in mind. All the user needs to provide is a csv file containing SMILES and their respective docking scores, the system then outputs a model that the user can use for the prediction of docking score for a new molecule.

Effect of Guided Discovery Instructional Strategy on the Academic Achievement and Retention among Chemistry Students on Nce 1 in Volumetric Analysis in Federal College of Education (Fce) Obudu

The study investigated the effect of the Guided Discovery Instructional Strategy on academic Achievement and Retention among Chemistry Students of NCE year 1 in volumetric analysis in F.C.E Obudu. Quasi-experimental Pretest-Posttest was the design used. The population of the study was made up of both male and female students totaling five hundred and twenty (520). The instrument used was a designed test questions on Chemistry Volumetric Analysis Test (CVAT). The test items were validated by experts in test and measurement Department in the Faculty of Education UNICAL and Chemistry Department, F.C.E Obudu. A reliability coefficient of 0.65 using Cronbach Alpha was used. Four research questions were raised and answered using mean and standard deviation statistical method. The results showed that, in the identical circumstances and settings, students instructed via the guided discovery method performed better than those taught using the traditional lecture style. The study also showed that when educating both sexes utilizing the Guided Discovery technique, there is no discernible difference in the academic performance of males and females. The study also shows that there was an important distinction in retention ability between the conventional lecture approach and the guided discovery method for all students. Result of the study showed that guided Discovery method is preferable in teaching Sciences (Chemistry) than lecture method in volumetric analysis. Recommendations were made showing clearly that the Guided Discovery method should be applied by science teachers in Nigeria because, it is effective in achievement and retention by students and also that, it does not create disparity in gender level of achievement and retention, thus it should be enshrined in the Science curriculum of studies in Tertiary Institutions in Nigeria.

Decoding the anti-cancer potential of Pexidartinib (PLX3397), a Fms-like tyrosine kinase 3 inhibitor, using next-generation knowledge discovery methods.

Acute Myeloid Leukemia (AML) is a complex hematologic malignancy characterized by the rapid proliferation of abnormal myeloid precursor cells. The FMS-like tyrosine kinase 3 (FLT3), a receptor tyrosine kinase, plays a pivotal role in regulating cell survival, proliferation, and differentiation within the hematopoietic system. Mutations in FLT3, particularly internal tandem duplications (ITDs) and point mutations within the tyrosine kinase domain (TKD), are prevalent in AML and are associated with poor prognosis and increased risk of relapse. The development of targeted therapies has revolutionized the landscape of cancer treatment by focusing on the inhibition of kinase signalling. Small-molecule inhibitors designed to selectively target receptor tyrosine kinases, such as PLX3397, have shown promising results in preclinical studies and early phase clinical trials. PLX3397 exerts its inhibitory effects by targeting CSF1R and KIT, leading to the disruption of receptor tyrosine kinase signalling cascades, suppression of leukemic cell growth, and induction of apoptosis. This study emphasizes the significance of FLT3 as a receptor tyrosine kinase as a therapeutic target for PLX3397. After evaluating the usefulness of PLX3397 as an enzyme inhibitor using ADMET prediction, PLX3397 was prepared for molecular docking in the FLT3 crystal structure (PDB: 4XUF). A molecular dynamics simulation was performed on PLX3397 to evaluate its binding affinity and protein stability in a simulated physiological environment. In conclusion, targeting FLT3 as a receptor tyrosine kinase with PLX3397 represents a promising therapeutic strategy for improving outcomes in patients with FLT3-mutated AML. Further clinical investigations are warranted to validate the efficacy and safety of PLX3397 and to optimize treatment strategies for AML patients based on the FLT3 mutational status.

Memahami Konsep Keselamatan dan Filsafat Kristen Tinjauan Dari Perspektif Teologis Dan Filosofi

On the other hand, from a Christian philosophical perspective, the concept of salvation can be approached through rational thinking and conceptual analysis. Christian philosophy often explores ontological and epistemological questions about the nature of salvation, such as the relationship between human freedom and God's sovereignty, as well as the nature of spiritual truth. This research highlights the differences in sources of authority, methods of truth discovery, and ultimate goals between theological and philosophical approaches to salvation. While theology relies on religious teachings, divine revelation, and religious experience, Christian philosophy tends to use logical reasoning and conceptual analysis. However, these two fields can complement each other in deepening understanding of the concept of salvation, creating creative and in-depth dialogue about the nature of the Christian faith. This research provides valuable insights for theologians, philosophers, and practitioners of Christianity to explore the concept of salvation holistically, combining deep theological understanding with critical philosophical analysis.

Probabilistic grammars for modeling dynamical systems from coarse, noisy, and partial data

Ordinary differential equations (ODEs) are a widely used formalism for the mathematical modeling of dynamical systems, a task omnipresent in scientific domains. The paper introduces a novel method for inferring ODEs from data, which extends ProGED, a method for equation discovery that allows users to formalize domain-specific knowledge as probabilistic context-free grammars and use it for constraining the space of candidate equations. The extended method can discover ODEs from partial observations of dynamical systems, where only a subset of state variables can be observed. To evaluate the performance of the newly proposed method, we perform a systematic empirical comparison with alternative state-of-the-art methods for equation discovery and system identification from complete and partial observations. The comparison uses Dynobench, a set of ten dynamical systems that extends the standard Strogatz benchmark. We compare the ability of the considered methods to reconstruct the known ODEs from synthetic data simulated at different temporal resolutions. We also consider data with different levels of noise, i.e., signal-to-noise ratios. The improved ProGED compares favourably to state-of-the-art methods for inferring ODEs from data regarding reconstruction abilities and robustness to data coarseness, noise, and completeness.

Computational Strategies in Drug Discovery: Leveraging Subtractive Genomic Analysis for Target Identification.

The utilization of In-silico subtractive genomic analysis has emerged as an important and essential method in modern drug discovery and development since it significantly improves the process of identifying and validating potential targets for discovering novel therapeutic compounds to treat severe infections caused by Antimicrobial-resistant (AMR) - pathogenic species. This review provides a complete overview of the methodology, advantages, disadvantages, and prospects, associated with subtractive genomic research in the context of drug discovery and development. The initial phase of analysis encompasses the retrieval of data, which serves as a foundation for the subsequent data mining process in Phase 1. After data mining, Phase 2 utilizes subtractive channels for the target's non-homology and essentiality analysis. Phase 3 of the study aims to provide a comprehensive understanding of prospective targets by their qualitative characterization. Further, Phase 4 of the study emphasizes on conducting structure-based analyses, which involves the determination, refinement, evaluation, and validation of three-dimensional structures of the target proteins, along with their active site prediction and selection of the novel therapeutic compounds against that active site on the obtained targets through virtual screening and docking studies by utilizing various databases and servers. The therapeutic compounds obtained can be then validated by in vitro and in vivo testing, thereby establishing a connection between the computational predictions and real applications.

Sparse dynamic discretization discovery via arc-dependent time discretizations

While many time-dependent network design problems can be formulated as time-indexed formulations with strong relaxations, the size of these formulations depends on the discretization of the time horizon and can become prohibitively large. The recently-developed dynamic discretization discovery (DDD) method allows many time-dependent problems to become more tractable by iteratively solving instances of the problem on smaller networks where each node has its own discrete set of departure times. However, in the current implementation of DDD, all arcs departing a common node share the same set of departure times. This causes DDD to be ineffective for solving problems where all near-optimal solutions require many distinct departure times at the majority of the high-degree nodes in the network. Region-based networks are one such structure that often leads to many high-degree nodes, and their increasing popularity underscores the importance of tailoring solution methods for these networks.To improve methods for solving problems that require many departure times at nodes, we develop a DDD framework where the set of departure times is determined on the arc level rather than the node level. We apply this arc-based DDD method to a temporal variant of the service network design problem (SND). We show that an arc-based approach is particularly advantageous when instances arise from region-based networks, and when candidate paths are fixed in the base graph for each commodity. Moreover, our algorithm builds upon the existing DDD framework and achieves these improvements with only benign modifications to the original implementation.

A novel rule mining method for knowledge discovery of relationships among indoor air quality, HVAC operation and occupants’ activities

This paper presents a novel data mining method to discover the occurrence time, co-occurrence and sequential association rules among indoor air quality (IAQ), Heating, Ventilation and Air Conditioning (HVAC) operation and occupants' activities. The novelty of this work includes a peak detection method to discover the pollutants' peaks and a rule mining method with extended rule selection criteria that considers both the time lags between events and relationships among multiple events. The peak detection method aimed to use moving windows and criteria to find the candidate peaks and the non-maximum suppression to filter the overlapped candidate peaks. The proposed rule mining method is based on the Apriori and Apriori-All with the rule selection criteria extended by labeling the occurrence time to events to find the rules among various occurrence times and multiple events and by limiting the time lags among the occurrences of the events to discover the sequential rules. Two datasets, one from 70 houses and one from a single commercial building, were used to evaluate the performance of this proposed data-driven method. The results showed that 93.19 % of the peak detection results matched the peaks detected from the data records. The CO2 peaked frequently at 12:00, 17:00 and from 19:00 to 24:00. The occurrence of occupants' activities could lead to pollutants’ concentrations and pollutants peaks within 2 hours in the residential buildings. The occurrence of a high Wifi connection number was frequently followed by a CO2 peak within 2 hours in the commercial building. These findings can support the development of control strategies for building HVAC systems to achieve better indoor air quality control.

Multi Methods of Teaching the Lord Jesus: A Strategy to Shape the Scientific Attitude of Students

Scientific attitude is a very important attitude to be possessed by every student in following the learning process. Therefore, this study aims to examine the teaching method of the Lord Jesus as a strategy in shaping the scientific attitude of students. Through a qualitative approach with literature review techniques, the author examines in depth the method of teaching Jesus as a strategy in shaping the scientific attitudes of students. The results of this study found that to shape the scientific attitude of students, it is necessary to be guided by the teaching methods of the Lord Jesus, such as discovery and inquary methods, discussion methods, problem-based, and storytelling methods. Therefore, teachers of Christian religious education are required to always emulate the Lord Jesus and think creatively in shaping the scientific attitude of students in accordance with teaching conditions in the 21st century.

Causal Analysis Discovers an Enhanced Impact of Tropical Western Pacific on Indian Summer Monsoon Subseasonal Anomalies

AbstractExisting studies have shown changes in the impact of atmospheric teleconnections on Indian Summer Monsoon (ISM) at interannual time scales due to the changing background state. However, the exploration of potential changes at subseasonal time scales remains limited. In this study, we use a causal discovery method to find the tropical atmospheric drivers of ISM subseasonal anomalies, and quantify the causal effects of the drivers on ISM, as well as the changes in causal effects over the past 40 years. We find the impact of tropical western Pacific on ISM subseasonal anomalies has strengthened, while the ISM self‐feedback has weakened. Wavenumber‐frequency analysis shows that the Rossby wave energy over 15°N–30°N at periods of 10–20 days has increased over the past four decades, which offers a plausible explanation for the enhanced tropical atmospheric teleconnection between western Pacific and ISM at subseasonal time scales.

Olaris Global Panel (OGP): A Highly Accurate and Reproducible Triple Quadrupole Mass Spectrometry-Based Metabolomics Method for Clinical Biomarker Discovery.

Mass spectrometry (MS)-based clinical metabolomics is very promising for the discovery of new biomarkers and diagnostics. However, poor data accuracy and reproducibility limit its true potential, especially when performing data analysis across multiple sample sets. While high-resolution mass spectrometry has gained considerable popularity for discovery metabolomics, triple quadrupole (QqQ) instruments offer several benefits for the measurement of known metabolites in clinical samples. These benefits include high sensitivity and a wide dynamic range. Here, we present the Olaris Global Panel (OGP), a HILIC LC-QqQ MS method for the comprehensive analysis of ~250 metabolites from all major metabolic pathways in clinical samples. For the development of this method, multiple HILIC columns and mobile phase conditions were compared, the robustness of the leading LC method assessed, and MS acquisition settings optimized for optimal data quality. Next, the effect of U-13C metabolite yeast extract spike-ins was assessed based on data accuracy and precision. The use of these U-13C-metabolites as internal standards improved the goodness of fit to a linear calibration curve from r2 < 0.75 for raw data to >0.90 for most metabolites across the entire clinical concentration range of urine samples. Median within-batch CVs for all metabolite ratios to internal standards were consistently lower than 7% and less than 10% across batches that were acquired over a six-month period. Finally, the robustness of the OGP method, and its ability to identify biomarkers, was confirmed using a large sample set.

Synthesis and interest in medicinal chemistry of β-phenylalanine derivatives (β-PAD): an update (2010-2022).

β-Phenylalanine derivatives (β-PAD) represent a structural family of therapeutic interest, either as components of drugs or as starting materials for access to key compounds. As scaffolds for medicinal chemistry work, β-PAD offer the advantage of great diversity and modularity, a chiral pseudopeptidic character that opens up the capacity to be recognized by natural systems, and greater stability than natural α-amino acids. Nevertheless, their synthesis remains a challenge in drug discovery and numerous methods have been devoted to their preparation. This review is an update of the access routes to β-PAD and their various therapeutic applications.

Performance Comparison of Machine Learning Methods in Discovery of BACE-1 Inhibitors in Alzheimer's Disease Therapy

Performance Comparison of Machine Learning Methods in Discovery of BACE-1 Inhibitors in Alzheimer's Disease Therapy