Improving classical scoring functions using random forest: The non‐additivity of free energy terms’ contributions in binding

Afifi, Karim; Al-Sadek, Ahmed Farouk

doi:10.1111/cbdd.13206

Cited by 30 publications

(24 citation statements)

References 21 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5. Further, to have a better understanding of the performance of our models, we have systematically compare our predictions with the state-of-the-art results in literatures, 1,4,15,15,19,21,22,40,41,50,55 as far as we known. The results are illustrated in Figs.…”

Section: Basic Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Persistent spectral based machine learning (PerSpect ML) for drug design

Meng,

Xia

2020

Preprint

View full text Add to dashboard Cite

In this paper, we propose persistent spectral based machine learning (PerSpect ML) models for drug design. Persistent spectral models, including persistent spectral graph, persistent spectral simplicial complex and persistent spectral hypergraph, are proposed based on spectral graph theory, spectral simplicial complex theory and spectral hypergraph theory, respectively. Different from all previous spectral models, a filtration process, as proposed in persistent homology, is introduced to generate multiscale spectral models. More specifically, from the filtration process, a series of nested topological representations, i,e., graphs, simplicial complexes, and hypergraphs, can be systematically generated and their spectral information can be obtained. Persistent spectral variables are defined as the function of spectral variables over the filtration value. Mathematically, persistent multiplicity (of zero eigenvalues) is exactly the persistent Betti number (or Betti curve). We consider 11 persistent spectral variables and use them as the feature for machine learning models in protein-ligand binding affinity prediction. We systematically test our models on three most commonly-used databases, including PDBbind-2007, PDBbind-2013 and PDBbind-2016 Our results, for all these databases, are better than all existing models, as far as we know. This demonstrates the great power of our PerSpect ML in molecular data analysis and drug design.

show abstract

Section: Basic Resultsmentioning

confidence: 99%

“…It has been found that our PerSpect ML model has delivered the best results in terms of both PCC and RMSE in all the three test cases. 1,4,15,15,19,21,22,40,41,50,55 2 Theory and methods…”

Section: Introductionmentioning

confidence: 99%

Persistent spectral based machine learning (PerSpect ML) for drug design

Meng,

Xia

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Compared to “classical” machine learning scoring functions, our method performs similarly to RF Score and other RF-based scoring functions [ 11 , 77 ]. Despite recent advances in deep learning architectures, which consistently outperform “classical” machine learning algorithms in image recognition and natural language processing [ 15 – 19 ], RFs remain very competitive for binding affinity predictions.…”

Section: Discussionmentioning

confidence: 99%

Learning protein-ligand binding affinity with atomic environment vectors

et al. 2021

View full text Add to dashboard Cite

Scoring functions for the prediction of protein-ligand binding affinity have seen renewed interest in recent years when novel machine learning and deep learning methods started to consistently outperform classical scoring functions. Here we explore the use of atomic environment vectors (AEVs) and feed-forward neural networks, the building blocks of several neural network potentials, for the prediction of protein-ligand binding affinity. The AEV-based scoring function, which we term AEScore, is shown to perform as well or better than other state-of-the-art scoring functions on binding affinity prediction, with an RMSE of 1.22 pK units and a Pearson’s correlation coefficient of 0.83 for the CASF-2016 benchmark. However, AEScore does not perform as well in docking and virtual screening tasks, for which it has not been explicitly trained. Therefore, we show that the model can be combined with the classical scoring function AutoDock Vina in the context of $$\Delta$$ Δ -learning, where corrections to the AutoDock Vina scoring function are learned instead of the protein-ligand binding affinity itself. Combined with AutoDock Vina, $$\Delta$$ Δ -AEScore has an RMSE of 1.32 pK units and a Pearson’s correlation coefficient of 0.80 on the CASF-2016 benchmark, while retaining the docking and screening power of the underlying classical scoring function.

show abstract

“…75 S3, together with references for all the different methods. 4,11,[20][21][22][23][24]51,63,64,[71][72][73][74] We also showed that the AEScore model presented here can be exploited in tandem with standard docking scoring functions using a ∆-learning approach, in order to improve the performance in docking and virtual screening (in which AEScore does not perform well).…”

Section: Supporting Informationmentioning

confidence: 90%

“…Compared to "classical" machine learning scoring functions, our method performs similarly to RF Score and other RF-based scoring functions. 11,73 Despite recent advances in deep learning architectures, which consistently outperform "classical" machine learning algorithms in image recognition and natural language processing, [15][16][17][18][19] RFs remain very competitive for binding affinity predictions. All top-performing machine learning and deep learning methods considered here achieve similar performance on the CASF benchmarks-as measured by…”

Section: Visualizationmentioning

confidence: 99%

Learning Protein-Ligand Binding Affinity with Atomic Environment Vectors

Meli¹,

Anighoro²,

Bodkin³

et al. 2020

Preprint

View full text Add to dashboard Cite

<div> <div> <div> <p>Scoring functions for the prediction of protein-ligand binding affinity have seen renewed interest in recent years when novel machine learning and deep learning methods started to consistently outperform classical scoring functions. Here we explore the use of atomic environment vectors (AEVs) and feed-forward neural networks, the building blocks of several neural network potentials, for the prediction of protein-ligand binding affinity. The AEV-based scoring function, which we term AEScore, is shown to perform as well or better than other state-of-the-art scoring functions on binding affinity prediction, with an RMSE of 1.22 pK units and a Pearson’s correlation coefficient of 0.83 for the CASF-2016 benchmark. However, AEScore does not perform as well in docking and virtual screening tasks. We therefore show that the model can be combined with the classical scoring function AutoDock Vina in the context of ∆-learning, where corrections to the AutoDock Vina scoring function are learned instead of the protein-ligand binding affinity itself. Combined with AutoDock Vina, ∆-AEScore has an RMSE of 1.32 pK units and a Pearson’s correlation coefficient of 0.80 on the CASF-2016 benchmark, while retaining the good docking and screening power of the underlying classical scoring function. </p> </div> </div> </div>

show abstract

Improving classical scoring functions using random forest: The non‐additivity of free energy terms’ contributions in binding

Cited by 30 publications

References 21 publications

Persistent spectral based machine learning (PerSpect ML) for drug design

Persistent spectral based machine learning (PerSpect ML) for drug design

Learning protein-ligand binding affinity with atomic environment vectors

Learning Protein-Ligand Binding Affinity with Atomic Environment Vectors

Contact Info

Product

Resources

About