A blind SAMPL6 challenge: insight into the octanol-water partition coefficients of drug-like molecules via a DFT approach

Arslan, Evrim; Fındık, Basak Koca; Avi̇yente, Vi̇ktorya

doi:10.1007/s10822-020-00284-3

Cited by 15 publications

(16 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ML model trained only for antimicrobial activity could not, however, recognize that the guideline of high lipophilicity is in practice bound by other antibiotic development criteria, e.g., cytotoxicity to human cells. This is because too high lipophilicity tends to lead to less targeted effects in the human body . The case of ALOGP exemplifies that when investigated from the viewpoint of accelerating antibiotics development pipelines, ideal ML models do not only max out a single antibiotic development criterion but guide efforts toward candidates that fulfill several criteria adequately.…”

Section: Results and Discussionmentioning

confidence: 99%

“…This is because too high lipophilicity tends to lead to less targeted effects in the human body. 44 The case of ALOGP exemplifies that when investigated from the viewpoint of accelerating antibiotics development pipelines, ideal ML models do not only max out a single antibiotic development criterion but guide efforts toward candidates that fulfill several criteria adequately. Therefore, continued efforts to develop reliable multitask models 45−49 are valuable even though they Comparison of Molecular Representation−Machine Learning Model Pairs.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Predicting Antimicrobial Activity of Conjugated Oligoelectrolyte Molecules via Machine Learning

et al. 2021

View full text Add to dashboard Cite

New antibiotics are needed to battle growing antibiotic resistance, but the development process from hit, to lead, and ultimately to a useful drug takes decades. Although progress in molecular property prediction using machine-learning methods has opened up new pathways for aiding the antibiotics development process, many existing solutions rely on large data sets and finding structural similarities to existing antibiotics. Challenges remain in modeling unconventional antibiotic classes that are drawing increasing research attention. In response, we developed an antimicrobial activity prediction model for conjugated oligoelectrolyte molecules, a new class of antibiotics that lacks extensive prior structure−activity relationship studies. Our approach enables us to predict the minimum inhibitory concentration for E. coli K12, with 21 molecular descriptors selected by recursive elimination from a set of 5305 descriptors. This predictive model achieves an R 2 of 0.65 with no prior knowledge of the underlying mechanism. We find the molecular representation optimum for the domain is the key to good predictions of antimicrobial activity. In the case of conjugated oligoelectrolytes, a representation reflecting the three-dimensional shape of the molecules is most critical. Although it is demonstrated with a specific example of conjugated oligoelectrolytes, our proposed approach for creating the predictive model can be readily adapted to other novel antibiotic candidate domains.

show abstract

Section: Results and Discussionmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Predicting Antimicrobial Activity of Conjugated Oligoelectrolyte Molecules via Machine Learning

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Singh recently reported a short review focused on the applications of DFT for drug design based on the investigation of drug-biomembrane interaction, aimed at proposing structure changes to improve biomembrane crossing and to enhance the drug efficacy profile [99]. Arslan et al [100] obtained, by DFT calculations, the octanol-water partition coefficients of drug-like molecules, with the aim to compare the Gibbs solvation free energies of potential drug candidates with experimentally available LogP values. In detail, the n-octanol/water partition coefficient, LogP or log K ow , describes the drug hydrophobicity and membrane permeability, a theoretically determined basic property for drug design, using the combinations of three functionals, two basis sets, an implicit salvation method with one explicit water molecule and performing a conformational search by using the semi-empirical PM3 method.…”

Section: Dft Calculationsmentioning

confidence: 99%

Quantitative Analysis of the Interactions of Metal Complexes and Amphiphilic Systems: Calorimetric, Spectroscopic and Theoretical Aspects

et al. 2022

View full text Add to dashboard Cite

Metals and metal-based compounds have many implications in biological systems. They are involved in cellular functions, employed in the formation of metal-based drugs and present as pollutants in aqueous systems, with toxic effects for living organisms. Amphiphilic molecules also play important roles in the above bio-related fields as models of membranes, nanocarriers for drug delivery and bioremediating agents. Despite the interest in complex systems involving both metal species and surfactant aggregates, there is still insufficient knowledge regarding the quantitative aspects at the basis of their binding interactions, which are crucial for extensive comprehension of their behavior in solution. Only a few papers have reported quantitative analyses of the thermodynamic, kinetic, speciation and binding features of metal-based compounds and amphiphilic aggregates, and no literature review has yet addressed the quantitative study of these complexes. Here, we summarize and critically discuss the recent contributions to the quantitative investigation of the interactions of metal-based systems with assemblies made of amphiphilic molecules by calorimetric, spectrophotometric and computational techniques, emphasizing the unique picture and parameters that such an analytical approach may provide, to support a deep understanding and beneficial use of these systems for several applications.

show abstract

“…cytotoxicity to human cells. This is because too high lipophilicity tends to lead to less targeted effects in the human body 45 . The case of ALOGP exemplifies that when investigated from the viewpoint of accelerating antibiotics development pipelines, ideal ML models do not only max out a single antibiotic development criterion but guide efforts towards candidates that fulfill several criteria adequately.…”

Section: Analysis Of Molecular Fingerprint Optimized For Conjugated O...mentioning

confidence: 99%

Predicting antimicrobial activity of conjugated oligoelectrolyte molecules via machine learning

Tiihonen,

Cox-Vazquez,

Liang

et al. 2021

Preprint

View full text Add to dashboard Cite

New antibiotics are needed to battle growing antibiotic resistance, but the development process from hit, to lead, and ultimately to a useful drug, takes decades. Although progress in molecular property prediction using machine-learning methods has opened up new pathways for aiding the antibiotics development process, many existing solutions rely on large datasets and finding structural similarities to existing antibiotics.Challenges remain in modelling of unconventional antibiotics classes that are drawing increasing research attention. In response, we developed an antimicrobial activity prediction model for conjugated oligoelectrolyte molecules, a new class of antibiotics that lacks extensive prior structure-activity relationship studies. Our approach enables us to predict minimum inhibitory concentration for E. coli K12, with 21 molecular descriptors selected by recursive elimination from a set of 5,305 descriptors. This predictive model achieves an R 2 of 0.65 with no prior knowledge of the underlying mechanism. We find the molecular representation optimum for the domain is the key to good predictions of antimicrobial activity. In the case of conjugated oligoelectrolytes, a representation reflecting the 3-dimensional shape of the molecules is most critical. Although it is demonstrated with a specific example of conjugated oligoelectrolytes, our proposed approach for creating the predictive model can be readily adapted to other novel antibiotic candidate domains.

show abstract

A blind SAMPL6 challenge: insight into the octanol-water partition coefficients of drug-like molecules via a DFT approach

Cited by 15 publications

References 40 publications

Predicting Antimicrobial Activity of Conjugated Oligoelectrolyte Molecules via Machine Learning

Predicting Antimicrobial Activity of Conjugated Oligoelectrolyte Molecules via Machine Learning

Quantitative Analysis of the Interactions of Metal Complexes and Amphiphilic Systems: Calorimetric, Spectroscopic and Theoretical Aspects

Predicting antimicrobial activity of conjugated oligoelectrolyte molecules via machine learning

Contact Info

Product

Resources

About