Prediction of Physicochemical Parameters by Atomic Contributions

By S. Wildman, G. Crippen. 1999

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation (2018, background): “Note that both scores are calculated from empirical prediction models whose parameters are estimated from related datasets [41, 1], and these scores are widely used in previous molecule generation papers [9, 22, 4, 39, 27].”
Pybel: a Python wrapper for the OpenBabel cheminformatics toolkit (2008, methodology): “In relation to cheminformatics, OpenBabel supports SMARTS searching [3], molecular fingerprints [4] (both Daylight-type, and structural-key based), and includes group contribution descriptors for LogP [5], polar surface area (PSA) [6] and molar refractivity (MR) [5].”
Recipe for a General, Powerful, Scalable Graph Transformer (2022, methodology): “Having no gain on the ZINC dataset is expected since the task is a combination of the computed octanol-water partition coefficient (cLogP) [56] and the synthetic accessibility score (SA-score) [17], both of which only count occurrences of local sub-structures.”
Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models (2018, unknown): “Computed via RDKit’s Crippen [73] function.”
Fréchet ChemblNet Distance: A metric for generative models for molecules (2018, methodology): “We compare the FCD to three commonly reported metrics: mean logP [22], mean druglikeness [3] and the internal diversity score with Tanimoto distance [1] of the generated samples.”