A5018674739- Computational Drug Discovery Methods
- Machine Learning in Materials Science
- Chemical Synthesis and Analysis
- Analytical Chemistry and Chromatography
- Process Optimization and Integration
- Innovative Microfluidic and Catalytic Techniques Innovation
- Microbial Natural Products and Biosynthesis
- Chemistry and Chemical Engineering
- Protein Structure and Dynamics
- Viral Infectious Diseases and Gene Expression in Insects
- Bioinformatics and Genomic Networks
- Pharmacogenetics and Drug Metabolism
- Genetics, Bioinformatics, and Biomedical Research
- Microbial Metabolic Engineering and Bioproduction
- RNA and protein synthesis mechanisms
- Metabolomics and Mass Spectrometry Studies
- Advanced Multi-Objective Optimization Algorithms
- Scientific Computing and Data Management
- History and advancements in chemistry
- Image Processing and 3D Reconstruction
- Advanced Scientific Research Methods
- Ubiquitin and proteasome pathways
- Research Data Management Practices
- Online Learning and Analytics
- Gene expression and cancer classification
Novo Nordisk (United States)
2024
Recursion (United States)
2024
Eli Lilly (United States)
2013-2022
University of Cyprus
2007-2015
Noesis Solutions (Belgium)
2010
Cyprus Institute
2009
Ludwig-Maximilians-Universität München
2009
Florida State University
2003
Artificial intelligence and machine learning have demonstrated their potential role in predictive chemistry synthetic planning of small molecules; there are at least a few reports companies employing
Venturing into the immensity of small molecule universe to identify novel chemical structure is a much discussed objective many methods proposed by chemoinformatics community. To this end, numerous approaches using techniques from fields computational de novo design, virtual screening and reaction informatics, among others, have been proposed. Although in principle commendable, practice there are several obstacles useful exploitation space. Prime them sheer number theoretically feasible...
The need for synthetic route design arises frequently in discovery-oriented chemistry organizations. While traditionally finding solutions to this problem has been the domain of human experts, several computational approaches, aided by algorithmic advances and availability large reaction collections, have recently reported. Herein we present our own implementation a retrosynthetic analysis method demonstrate its capabilities an attempt identify routes collection approved drugs. Our results...
Drug discovery and development is a complex, lengthy process, failure of candidate molecule can occur as result combination reasons, such poor pharmacokinetics, lack efficacy, or toxicity. Successful drug candidates necessarily represent compromise between the numerous, sometimes competing objectives so that benefits to patients outweigh potential drawbacks risks. De novo design involves searching an immense space feasible, druglike molecules select those with highest chances becoming drugs...
Deep learning has drawn significant attention in different areas including drug discovery. It been proposed that it could outperform other machine algorithms, especially with big data sets. In the field of pharmaceutical industry, models are built to understand quantitative structure–activity relationships (QSARs) and predict molecular activities, absorption, distribution, metabolism, excretion (ADME) properties, using only structures. Previous reports have demonstrated advantages deep...
DNA-encoded library (DEL) technology is a novel ligand identification strategy that allows the synthesis and screening of unprecedented chemical diversity more efficiently than conventional methods. However, no reports have been published to systematically study how increase improve molecular property space can be covered with DEL. This report describes development application eDESIGNER, an algorithm comprehensively generates all possible designs, enumerates profiles samples from each...
Increasing the success rate and throughput of drug discovery will require efficiency improvements throughout process that is currently used in pharmaceutical community, including crucial step identifying hit compounds to act as drivers for subsequent optimization. Hit identification can be carried out through large compound collection screening often involves generation testing many hypotheses based on available knowledge. In practice, hypothesis involve selection promising chemical...
Modern drug discovery is an iterative process relying on hypothesis generation through exploitation of available data and testing that produces informative results necessary for subsequent rounds exploration. In this setting, consists designing chemical structures likely to meet the pharmaceutically relevant objectives project pursued while involves compound synthesis biological assays query hypothesis. While much attention has been placed effective design, it often case efforts lead novel...
Fragment-based ligand discovery has been successful in targeting diverse proteins. Despite drug-like molecules having more 3D character, traditional fragment libraries are largely composed of flat, aromatic fragments. The use 3D-enriched fragments for enhancing library diversity is underexplored especially against protein-protein interactions. Here, we evaluate using bromodomains. Bromodomains highly ligandable, but selectivity remains challenging, particularly bromodomain and extraterminal...
Hierarchical clustering algorithms such as Wards or complete-link are commonly used in compound selection and diversity analysis. Many applications utilize binary representations of chemical structures, MACCS keys Daylight fingerprints, dissimilarity measures, the Euclidean Soergel measure. However, hierarchical can generate ambiguous results owing to what is known cluster analysis literature ties proximity problem, i.e., compounds clusters that equidistant from a given collection. Ambiguous...
As the use of high-throughput screening systems becomes more routine in drug discovery process, there is an increasing need for fast and reliable analysis massive amounts resulting data. At forefront methods used data reduction, often assisted by cluster analysis. Activity thresholds reduce set under investigation to manageable sizes while clustering enables detection natural groups that reduced subset, thereby revealing families compounds exhibit increased activity toward a specific...
Models that accurately predict properties based on chemical structure are valuable tools in the sciences. However, for many properties, public and private training sets typically small, making it difficult models to generalize well outside of data. Recently, this lack generalization has been mitigated by using self-supervised pretraining large unlabeled datasets, followed finetuning smaller, labeled datasets. Inspired these advances, we report MolE, a Transformer architecture adapted...
Could high-quality in silico predictions drug discovery eventually replace part or most of experimental testing? To evaluate the agreement selectivity data from different predictive sources, we introduce new metric concordance minimum significant ratio (cMSR). Empowered by cMSR, find overall level between predicted and to be comparable that found results sources. However, for molecules are either highly selective potent, sources is significantly higher than values. We also show computational...
Educators participating in networked learning communities have very little support from integrated tools evaluating students’ activities flow and examining learners’ online behaviour. There is a need for non‐intrusive ways to monitor progress order better follow their process appraise the course effectiveness. This paper presents conceptual framework an innovative tool, called LMSAnalytics, that allows teachers evaluators easily track behaviour, make judgments about activity gain insight...
Models that accurately predict properties based on chemical structure are valuable tools in drug discovery. However, for many properties, public and private training sets typically small, it is difficult the models to generalize well outside of data. Recently, large language have addressed this problem by using self-supervised pretraining unlabeled datasets, followed fine-tuning smaller, labeled datasets. In paper, we report MolE, a molecular foundation model adapts DeBERTa architecture be...
A simple yet powerful programming tool enabling in silico experimentation, end-to-end data management through web services as well use of grid and cloud processing power is scientific workflows. This technology receiving considerable interest recent years primarily due to its ability promote support collaboration among large distributed research teams. The paper reviews the Scientific Workflows Management Systems (SWMS) field investigates detail popular open source workflow systems used...
Computer-aided drug discovery techniques have been widely used in recent years to support the development of new pharmaceuticals. Virtual screening, computational counterpart experimental attempts replicate results from vitro and vivo methods through use silico models tools. This paper presents LISIs platform; a web based scientific workflow system for virtual screening that has implemented primarily chemoprevention agents. We describe overall design as well implementation its various...