Discovering H-Bonding Rules in Crystals with Inductive Logic Programming

Ando, Howard Y.; Dehaspe, Luc; Luyten, Walter; Craenenbroeck, Elke Van; Vandecasteele, Henk; Meervelt, Luc Van

doi:10.1021/mp060034z

Cited by 16 publications

(23 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Note that diminishing marginal returns are to be expected as performance increases. The ROC 50 results confirm that the augmen- 6 Note that we have here abused the notation and consider e = 1−AUROC rather than e = 1−accuracy Note that the WDK results reported in Table 2 are worse than those obtained by Ceroni et al…”

Section: Resultssupporting

confidence: 63%

Molecular Graph Augmentation with Rings and Functional Groups

Grave

Costa

2010

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Molecular graphs are a compact representation of molecules, but may be too concise to obtain optimal generalization performance from graph-based machine learning algorithms. Over centuries, chemists have learned what are the important functional groups in molecules. This knowledge is normally not manifest in molecular graphs. In this paper, we introduce a simple method to incorporate this type of background knowledge: we insert additional vertices with corresponding edges for each functional group and ring structure identified in the molecule.We present experimental evidence that, on a wide range of ligand-based tasks and data sets, the proposed augmentation method improves the predictive performance over several graph kernel based QSAR models. When the augmentation technique is used with the recent Pairwise Maximal Common Subgraphs Kernel, we achieve a significant improvement over the current state-of-the-art on the NCI-60 cancer data set in 28 out of 60 cell lines, with the other 32 cell lines showing no significant difference in accuracy. Finally, on the Bursi mutagenicity data set, we obtain near-optimal predictions.

show abstract

Section: Resultssupporting

confidence: 63%

Molecular Graph Augmentation with Rings and Functional Groups

Grave

Costa

2010

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…−7.1 kcal/mol, was regarded as a reference value. The binding energy information together with other required chemical information was fed in the DMax Chemistry Assistant (DCA) software 18 to generate the inhibitor binding model. The generated inhibitor model suggested that compounds containing methyl, amide (thioamides, etc.…”

Section: Resultsmentioning

confidence: 99%

“…In this ligand-based VS study, inductive logic programming (ILP) was used to develop molecular search patterns, and molecular docking was performed to estimate the binding affinities of potential SIRT1 ligands. ILP can consider specific characteristics of compounds and human-generated rules as background knowledge to outperform traditional approaches 18 .…”

mentioning

confidence: 99%

Ligand-based virtual screening and inductive learning for identification of SIRT1 inhibitors in natural products

Sun

Zhou

Zhu

et al. 2016

Sci Rep

View full text Add to dashboard Cite

Sirtuin 1 (SIRT1) is a nicotinamide adenine dinucleotide-dependent deacetylase, and its dysregulation can lead to ageing, diabetes, and cancer. From 346 experimentally confirmed SIRT1 inhibitors, an inhibitor structure pattern was generated by inductive logic programming (ILP) with DMax Chemistry Assistant software. The pattern contained amide, amine, and hetero-aromatic five-membered rings, each of which had a hetero-atom and an unsubstituted atom at a distance of 2. According to this pattern, a ligand-based virtual screening of 1 444 880 active compounds from Chinese herbs identified 12 compounds as inhibitors of SIRT1. Three compounds (ZINC08790006, ZINC08792229, and ZINC08792355) had high affinity (−7.3, −7.8, and −8.6 kcal/mol, respectively) for SIRT1 as estimated by molecular docking software AutoDock Vina. This study demonstrated a use of ILP and background knowledge in machine learning to facilitate virtual screening.

show abstract

“…Listing 4 shows a kLog script for the Bursi domain. Relevant predicates in the extensional database are a/2, b/3 (atoms and bonds, respectively, extracted from the chemical structure), sub/3 (functional groups, computed by DMax Chemistry Assistant [55,56]), fused/3, connected/4 (direct connection between two functional groups), linked/4 (connection between functional groups via an aliphatic chain). Aromaticity (used in the bond-type property of b/3) was also computed by DMax Chemistry Assistant.…”

Section: Predicting a Single Property Of One Interpretationmentioning

confidence: 99%

kLog: A language for logical and relational learning with kernels

Costa

Raedt

Grave

2014

Artificial Intelligence

View full text Add to dashboard Cite

We introduce kLog, a novel approach to statistical relational learning. Unlike standard approaches, kLog does not represent a probability distribution directly. It is rather a language to perform kernel-based learning on expressive logical and relational representations. kLog allows users to specify learning problems declaratively. It builds on simple but powerful concepts: learning from interpretations, entity/relationship data modeling, logic programming, and deductive databases. Access by the kernel to the rich representation is mediated by a technique we call graphicalization: the relational representation is first transformed into a graph -in particular, a grounded entity/relationship diagram. Subsequently, a choice of graph kernel defines the feature space. kLog supports mixed numerical and symbolic data, as well as background knowledge in the form of Prolog or Datalog programs as in inductive logic programming systems. The kLog framework can be applied to tackle the same range of tasks that has made statistical relational learning so popular, including classification, regression, multitask learning, and collective classification. We also report about empirical comparisons, showing $ PF was a visiting professor at KU Leuven and FC a postdoctoral fellow at KU Leuven while this work was initiated * Corresponding author

show abstract

Discovering H-Bonding Rules in Crystals with Inductive Logic Programming

Cited by 16 publications

References 18 publications

Molecular Graph Augmentation with Rings and Functional Groups

Molecular Graph Augmentation with Rings and Functional Groups

Ligand-based virtual screening and inductive learning for identification of SIRT1 inhibitors in natural products

kLog: A language for logical and relational learning with kernels

Contact Info

Product

Resources

About