Dataset Information

Benchmarking the acceleration of materials discovery by sequential learning.

ABSTRACT: Sequential learning (SL) strategies, i.e. iteratively updating a machine learning model to guide experiments, have been proposed to significantly accelerate materials discovery and research. Applications on computational datasets and a handful of optimization experiments have demonstrated the promise of SL, motivating a quantitative evaluation of its ability to accelerate materials discovery, specifically in the case of physical experiments. The benchmarking effort in the present work quantifies the performance of SL algorithms with respect to a breadth of research goals: discovery of any "good" material, discovery of all "good" materials, and discovery of a model that accurately predicts the performance of new materials. To benchmark the effectiveness of different machine learning models against these goals, we use datasets in which the performance of all materials in the search space is known from high-throughput synthesis and electrochemistry experiments. Each dataset contains all pseudo-quaternary metal oxide combinations from a set of six elements (chemical space), the performance metric chosen is the electrocatalytic activity (overpotential) for the oxygen evolution reaction (OER). A diverse set of SL schemes is tested on four chemical spaces, each containing 2121 catalysts. The presented work suggests that research can be accelerated by up to a factor of 20 compared to random acquisition in specific scenarios. The results also show that certain choices of SL models are ill-suited for a given research goal resulting in substantial deceleration compared to random acquisition methods. The results provide quantitative guidance on how to tune an SL strategy for a given research goal and demonstrate the need for a new generation of materials-aware SL algorithms to further accelerate materials discovery.

SUBMITTER: Rohr B

PROVIDER: S-EPMC8157525 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:In recent years machine learning has transformed many aspects of the drug discovery process including small molecule design for which the prediction of the bioactivity is an integral part. Leveraging structural information about the interactions between a small molecule and its protein target has great potential for downstream machine learning scoring approaches, but is fundamentally limited by the accuracy with which protein:ligand complex structures can be predicted in a reliable and automated fashion. With the goal of finding practical approaches to generating useful kinase:inhibitor complex geometries for downstream machine learning scoring approaches, we present a kinase-centric docking benchmark assessing the performance of different classes of docking and pose selection strategies to assess how well experimentally observed binding modes are recapitulated in a realistic cross-docking scenario. The assembled benchmark data set focuses on the well-studied protein kinase family and comprises a subset of 589 protein structures co-crystallized with 423 ATP-competitive ligands. We find that the docking methods biased by the co-crystallized ligand-utilizing shape overlap with or without maximum common substructure matching-are more successful in recovering binding poses than standard physics-based docking alone. Also, docking into multiple structures significantly increases the chance to generate a low RMSD docking pose. Docking utilizing an approach that combines all three methods (Posit) into structures with the most similar co-crystallized ligands according to shape and electrostatics proofed to be the most efficient way to reproduce binding poses achieving a success rate of 66.9 % across all included systems. The studied docking and pose selection strategies-which utilize the OpenEye Toolkit-were implemented into pipelines of the KinoML framework allowing automated and reliable protein:ligand complex generation for future downstream machine learning tasks. Although focused on protein kinases, we believe the general findings can also be transferred to other protein families.

Project description:Liquid-filled porous materials exist widely in nature and engineering fields, with the diffusion of substances in them playing an important role in system functions. Although surface evaporation is often inevitable in practical scenarios, the evaporation effects on diffusion behavior in liquid-filled porous materials have not been well explored yet. In this work, we performed noninvasive diffusion imaging experiments to observe the diffusion process of erioglaucine disodium salt dye in a liquid-filled nitrocellulose membrane under a wide range of relative humidities (RHs). We found that evaporation can significantly accelerate the diffusion rate and alter concentration distribution compared with the case without evaporation. We explained the accelerated diffusion phenomenon by the mechanism that evaporation would induce a weak flow in liquid-filled porous materials, which leads to convective diffusion, i.e., evaporation-induced flow and diffusion (EIFD). Based on the EIFD mechanism, we proposed a convective diffusion model to quantitatively predict the diffusion process in liquid-filled porous materials under evaporation and experimentally validated the model. Introducing the dimensionless Peclet (P e) number to measure the relative contribution of the evaporation effect to pure molecular diffusion, we demonstrated that even at a high RH of 95%, where the evaporation effect is usually assumed negligible in common sense, the evaporation-induced diffusion still overwhelms the molecular diffusion. The flow velocity induced by evaporation in liquid-filled porous materials was found to be 0.4-5 μm/s, comparable to flow in many biological and biomedical systems. The present analysis may help to explain the driving mechanism of tissue perfusion and provide quantitative analysis or inspire new control methods of flow and material exchange in numerous cutting-edge technologies, such as paper-based diagnostics, hydrogel-based flexible electronics, evaporation-induced electricity generation, and seawater purification.

Dataset Information

Benchmarking the acceleration of materials discovery by sequential learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets