New machine learning and physics-based scoring functions for drug discovery

Guedes, Isabella Alvim; Barreto, André; Marinho, Diogo; Krempser, Eduardo; Kuenemann, Mélaine A.; Spérandio, Olivier; Dardenne, Laurent E.; Miteva, Maria A.

doi:10.1038/s41598-021-82410-1

Cited by 146 publications

(101 citation statements)

References 83 publications

Supporting

Mentioning

100

Contrasting

Order By: Relevance

“…However, the best score is achieved by the RF model in Fig. 4 c, which generally fits well in physics-induced problems 49 . Following the variable importance result obtained from Fig.…”

Section: Resultsmentioning

confidence: 62%

Nanoscale slip length prediction with machine learning tools

Sofos

Karakasidis

2021

Sci Rep

View full text Add to dashboard Cite

This work incorporates machine learning (ML) techniques, such as multivariate regression, the multi-layer perceptron, and random forest to predict the slip length at the nanoscale. Data points are collected both from our simulation data and data from the literature, and comprise Molecular Dynamics simulations of simple monoatomic, polar, and molecular liquids. Training and test points cover a wide range of input parameters which have been found to affect the slip length value, concerning dynamical and geometrical characteristics of the model, along with simulation parameters that constitute the simulation conditions. The aim of this work is to suggest an accurate and efficient procedure capable of reproducing physical properties, such as the slip length, acting parallel to simulation methods. Non-linear models, based on neural networks and decision trees, have been found to achieve better performance compared to linear regression methods. After the model is trained on representative simulation data, it is capable of accurately predicting the slip length values in regions between or in close proximity to the input data range, at the nanoscale. Results also reveal that, as channel dimensions increase, the slip length turns into a size-independent material property, affected mainly by wall roughness and wettability.

show abstract

“…However, the best score is achieved by the RF model in Fig. 4 c, which generally fits well in physics-induced problems 49 . Following the variable importance result obtained from Fig.…”

Section: Resultsmentioning

confidence: 62%

Nanoscale slip length prediction with machine learning tools

Sofos

Karakasidis

2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…However, in the present work, we were looking for the potential anti-cholesteremic activity of the compounds. Dockthor program is developed by GMMSB/LNCC group and it uses MMFF94S force field as scoring function (Guedes et al, 2021). It can easily dock highly flexible compounds with up to 40 rotatable bonds (Santos et al, 2020).…”

Section: Methodsmentioning

confidence: 99%

Molecular docking studies of Chenopodium album Linn with Lanosterol synthase enzyme

Madhuri¹,

Goteti

2021

JANS

View full text Add to dashboard Cite

Cardiovascular diseases (CVD) are the major cause of death among people across the globe. Hypercholesterolemia is one of the major contributing factors for CVD. Molecules that bind with Lanosterol synthase enzyme, can be potential drug targets. Statin group of compounds like Simvastatin, cerivastatin, Atorvastatin etc., used for treating hypercholesterolemia have side effects and hence there is a growing demand for plant derived flavonoids. This work focusses on studying the compounds quercetin-3-O-(2??,6??-di-O-?-l-rhamnopyranosyl)-?-d-glucopyranoside, kaempferol-3-O-(2??,6??-di-O-?-l-rhamnopyranosyl)-?-d-glucopyranoside, rutin; quercetin-3-O-?-d-glucopyranoside (Iso quercetin); and kaempferol-3-O-?-d-glucopyranoside (Astragalin) present in Chenopodium album Linn to inhibit Lanosterol synthase. Bioactivity score, drug likeness character was assessed in silico. Based on bioactivity spectrum, it is observed that the molecules are biologically active and the probability of these compounds to be biologically active is ranging from 0.784 to 0.992, suggesting that these compounds are effective for treating hypercholesterolemia. In the molecular docking studies, the compounds binding affinity score was in agreement that the molecules have the potential to be used as an alternative to the statin group of compounds in treating cholesterol.

show abstract

“…3D structure alignments were performed and visualized in PYMOL 2.2 (Schr ödinger, LLC.). Attempted docking studies with compound 2 were performed using the closed form homology model of YxeK with an aligned FMN cofactor from CmoJ using the DockThor webserver [44,45].…”

Section: Attempted Crystallization Conditions and Homology Modelmentioning

confidence: 99%

Bacterial flavoprotein monooxygenase YxeK salvages toxic S‐(2‐succino)‐adducts via oxygenolytic C–S bond cleavage

et al. 2021

View full text Add to dashboard Cite

Thiol-containing nucleophiles such as cysteine react spontaneously with the citric acid cycle intermediate fumarate to form S-(2-succino)-adducts. In Bacillus subtilis, a salvaging pathway encoded by the yxe operon has recently been identified for the detoxification and exploitation of these compounds as sulfur sources. This route involves acetylation of S-(2succino)cysteine to N-acetyl-2-succinocysteine, which is presumably converted to oxaloacetate and N-acetylcysteine, before a final deacetylation step affords cysteine. The critical oxidative cleavage of the C-S bond of N-acetyl-S-(2-succino)cysteine was proposed to depend on the predicted flavoprotein monooxygenase YxeK. Here, we characterize YxeK and verify its role in S-(2-succino)-adduct detoxification and sulfur metabolism. Detailed biochemical and mechanistic investigation of YxeK including 18 O-isotope-labeling experiments, homology modeling, substrate specificity tests, site-directed mutagenesis, and (pre-)steady-state kinetics provides insight into the enzyme's mechanism of action, which may involve a noncanonical flavin-N5-peroxide species for C-S bond oxygenolysis.

show abstract

New machine learning and physics-based scoring functions for drug discovery

Cited by 146 publications

References 83 publications

Nanoscale slip length prediction with machine learning tools

Nanoscale slip length prediction with machine learning tools

Molecular docking studies of Chenopodium album Linn with Lanosterol synthase enzyme

Bacterial flavoprotein monooxygenase YxeK salvages toxic S‐(2‐succino)‐adducts via oxygenolytic C–S bond cleavage

Contact Info

Product

Resources

About