Collision cross section prediction with molecular fingerprint using machine learning

HIGHLIGHTS

SUMMARY

The authors built a classification model to predict the chemical super classes based on their fingerprints. All 13,324 chemicals were combined and to build a direct CCS prediction model. L-tenuazonic acid (C10H15NO3) was predicted to have a twice higher CCS than the measured one by the class-based model (35% higher by the direct prediction model). Predicted CCS of vinyl acetate (C4H6O2) was 127.4 Å2 through the class-based model, and 147.9 Å2 by direct prediction, while the empirical one was 227.2 Å2. The authors applied direct CCS prediction . . .

 



If you want to have access to all the content you need to log in!

Thanks :)

Username or E-mail:

If you don't have an account, you can create one here.