Version 2023 of the Classic pKa calculator in Percepta includes a significantly expanded training set for improved prediction accuracy, broader chemical applicability; and algorithmic changes that deliver impressive speed enhancements to this industry-leading software.
26 years after the release of their first ionization predictor, ACD/Labs, an informatics company that develops and commercializes software in support of R&D, today announces the release of their latest version of molecular property calculators on the Percepta Platform.
Andrew Anderson, Vice President of Innovation and Informatics Strategy at ACD/Labs says, Metaphorically speaking, data is the 'heartbeat' of chemistry R&D, generated during either physical experimentation or in silico calculation. In an age where research groups are looking to harness the power of data in artificial intelligence and machine learning frameworks, prediction reliability is one of the most important success determinants. Our new version of Percepta pKa exemplifies our commitment to enabling reliable, data-driven R&D. Many molecules under investigation in R&D labs are susceptible to ionization. Knowing 'when and where' is of paramount importance since the same chemical might behave differently under different conditions. Reliable and accurate predictions not only help computational scientists develop accurate SAR models; these predictions also help scientists account for the ionization properties of molecules during their 'design, make, test' cycles. Finally, accurate descriptor prediction will empower data scientists with reliable insights for the 'next crop' of ground-breaking innovation activity. We're proud to support our large user community with this important new release.
Ionizationfrequently expressed as the acid dissociation constant for molecules with ionizable centers, or pKais a fundamental property that affects the behavior of chemical compounds in environmental, pharmaceutical, agrochemical, and other fields. It plays a crucial role in defining other properties like lipophilicity (logD), aqueous solubility, toxicity (hERG), and helps with understanding the chemical reactivity of molecular entities. In biopharma, it is used to model more complex ADME behaviors such as absorption, bioavailability, and clearance, and to develop delivery systems and formulations for new drug entities. pKa values are used by researchers to understand and modulate the behavior of molecules in biological and environmental systems, and to successfully apply separation technology such as chromatography. Furthermore, predictive software is an integral part of green chemistry initiatives and sustainable practices, providing a reliable alternative to physical experiments.
This new release is a shining example of our continued commitment to excellence, said Andrius Sazonovas, Head of Percepta Software Development at ACD/Labs. Since the introduction of our first ionization predictor in 1997; my colleagues and I have overseen the improvements to our computational software and have worked on numerous customer projects to apply our modeling methodology to proprietary datasets at innovative chemistry-driven organizations. In our tests of Percepta version 2023, we saw a 510 fold increase in the speed of calculation of pKa for various datasets. Significant improvement was also observed for prediction accuracy. While this varies for different datasets, 120% improvement was noted for a set of ~370 novel pharmaceutical compounds with ~600 pKa values in a recent collaborative project. This release truly represents the next generation of pKa prediction from ACD/Labs, with respect to performance, reliability, and speed.
For more information about in silico predictions available from ACD/Labs, meet our team at Booth 933 in the ACS exhibition hall or visit acdlabs.com/percepta.
ACD/Labs is a leading provider of scientific software for R&D. We help customers in >94 countries around the world assemble digitized analytical, structural, and molecular information for effective decision-making, problem solving, and product lifecycle control. Our enterprise technologies enable automation of molecular characterization and facilitate chemically intelligent knowledge management. ACD/Labs provides worldwide sales and support and brings decades of experience and success helping organizations innovate and create efficiencies in their workflows. For more information, please visit www.acdlabs.com. Follow us on Twitter and LinkedIn.
View source version on newsdirect.com: https://newsdirect.com/news/a-leap-in-ionization-prediction-acd-labs-unveils-groundbreaking-version-of-modeling-software-at-acs-fall-2023-164276509
Disclaimer: The views, suggestions, and opinions expressed here are the sole responsibility of the experts. No Gio News UK journalist was involved in the writing and production of this article.