Machine Learning Algorithm        

CCS-Predict Pro

Machine learning power, at your fingertips. Accurate predictions, instantly

The Next-Generation Collision Cross Section Predictor for Metabolomics and Lipidomics

CCS-Predict Pro is a machine learning algorithm that accurately predicts collision cross section (CCS) values from 2D compound structures. It is faster, more integrated, and more accessible than traditional CCS prediction methods. CCS-Predict Pro can be used to improve compound annotation, discriminate between isomers, identify metabolites in complex samples, and characterize lipids. It is integrated with MetaboScape®, a leading software platform for metabolomics and lipidomics data analysis. 

CCS-Predict Pro is a valuable tool for researchers at all levels, helping them to improve the accuracy, efficiency, and scope of their research.

Accurate and reproducible CCS predictions
CCS-Predict Pro uses machine learning to predict CCS values with high accuracy and reproducibility, even for compounds that are not present in existing CCS libraries.
Discrimination between isomers
In combination with ion mobility separation, CCS-Predict Pro can be used to distinguish between isomeric compounds, which can be challenging using MS and MS/MS data alone. 
Improved compound annotation
CCS-Predict Pro can be used to improve the accuracy and efficiency of compound annotation by providing an additional orthogonal criterion to MS and MS/MS data. 
Seamless software integration
CCS-Predict Pro integrates seamlessly with MetaboScape®, a leading software platform for metabolomics and lipidomics data analysis.

Confident annotations made easy

Metabolomics and lipidomics researchers have long struggled with the challenge of using retention time as an orthogonal criterion for compound annotations. Retention times often vary across labs, batches, instruments, columns, and even users' favored methods, making them difficult to use for high-throughput or confident annotations. CCS-Predict Pro is a game-changer that introduces timsCCS values into your annotation workflow. This distinct measurement considers the size and shape of molecules, providing unprecedented annotation accuracy. Plus, CCS is reproducible across instruments, labs, and ionization sources. Even if you don't have an experimental CCS value for a compound, CCS-Predict Pro's highly accurate machine learning algorithm can predict CCS values on the fly.

No more limitations with CCS libraries

Traditional CCS libraries are valuable, but they cannot keep up with the ever-expanding chemical space of metabolites and lipids. This means that researchers are often unable to find CCS values for the compounds they are studying, which can limit their ability to annotate compounds accurately. 

CCS-Predict Pro overcomes this limitation by accurately predicting CCS values directly from easily accessible 2D compound structures. This is a significant advantage, as researchers can now annotate a wider range of compounds with greater confidence, even if they are not present in existing CCS libraries. 

Streamlined candidate ranking

When elucidating the structure of bioactive small molecules, researchers often generate a large number of candidate structures. This can make it difficult to identify the most likely structure, especially if the candidate structures are similar in size and shape. 

CCS-Predict Pro can be used to streamline the candidate ranking process by ranking candidate structures based on their predicted CCS values. This can help researchers to quickly and easily identify the most likely structure, saving time and resources. 

For example, imagine that a researcher is trying to identify the structure of a new metabolite in a blood sample. They have generated a number of candidate structures, but they are not sure which one is the most likely. 

Researchers can use CCS-Predict Pro to predict the CCS values for all the candidate structures and compare against actual features measured in a sample. CCS-Predict Pro will then rank candidate structures to features based on their measured CCS values. The researcher can then focus their efforts on the candidate structures with the best matching CCS value, helping to streamline the elucidation of the most likely structures.

Overall, CCS-Predict Pro is a powerful tool that can help metabolomics and lipidomics researchers to streamline the candidate ranking process and to make quicker, more confident decisions about which structures to pursue.