Power-Packed Molecular Property Predictors Get a Boost in Newest Version Release

19 Nov 2007
Greg Smith
Analyst / Analytical Chemist

A new logP prediction model, an enhanced pKa model, and a customizable report for quickly profiling individual compounds are just some of the highlights of ACD/Labs PhysChem Version 11.0

Advanced Chemistry Development, Inc., (ACD/Labs) announces the release of our latest version of ACD/Labs PhysChem software. Version 11 provides updated molecular property predictions for a broader variety of chemical classes through enhanced models; and offers a new customizable interface for instant interactive review of results.

In version 11 we announce a new logP prediction model based on experimental data from more than 25,000 compounds. Substantial increase of the internal database (the training set) from version 10, with experimental data, has expanded and diversified chemical space coverage, particularly for compounds of pharmaceutical interest. Enhancements to ACD/pKa DB come from the addition of over 2,500 new compounds and experimental data into the training database, resulting in increased chemical space coverage and greater prediction accuracy.

Since the PhysChem product line is heavily integrated, improvements to the logP and pKa models augment the prediction quality of our whole range of predictive software products. Our most popular applications that predict pH profiles of the distribution coefficient (logD) and aqueous solubility use logP and pKa predictions in the background and will therefore be influenced by the new and enhanced models.

To further accommodate novel or proprietary chemistry, version 11 brings improvements to our algorithm training capabilities to encourage more frequent use of in-house experimental data for improving prediction accuracy. This will help computational chemists ensure that quality predictive models are available for the proprietary chemical classes their researchers might be exploring in the future.

ACD/Quick Report is a new customizable feature of the software that conveniently displays a snapshot of molecular property profiles, one compound at a time, making ACD/Labs PhysChem software an ideal fit for the natural workflow of synthetic chemists.

The last 12 years of active development and refinement has helped ACD/Labs excel in the area of prediction accuracy, and offer intuitive tools to scientists. This year’s milestone release is the culmination of years of development work spearheaded by the ACD/Labs expert team and based on feedback from users. ACD/Labs remains committed to developing tools that aid in the research process, and that empower scientists with the information they need to reach their goals and achieve success.

Links

Tags