Hesham Ghobarah: My name is Hesham Ghobarah. I'm a pharmaceutical application scientist with AB SCIEX. And in this webinar, I will provide a short overview of the new functions in MetabolitePilot software version 1.5 and how it can help streamline both data analysis and interpretation. I will begin by reviewing the current bottlenecks in the Met ID workflow and where hardware and software solutions can help us to increase efficiency and speed. Then, I will discus how web capabilities, how MetabolitePilot software can help solve the bottlenecks in data processing and how advanced new features in version 1.5 will help address data interpretation in areas of structure elucidation, scan the inter sample correlation.

With advances in high resolution LCMS systems, large amounts of data can be acquired very quickly. And the amount of useful information that's contained in the data is higher. For example, with the Triple Soft 5600 system, high resolution full scan and high resolution product ion spectra can be acquired at a very high speed in the same run. This places more demands on data processing and interpretation. Two of the more time consuming aspects of data interpretation are the structure elucidation of metabolites and the correlation of metabolism data for the same parent compound across multiple samples or multiple species.

The Triple Soft 5600 system combined with MetabolitePilot software offers advanced features at every step of the Meta ID process. In today's webinar, I will focus on how MetabolitePilot version 1.5 can help streamline the data interpretation. I will also review its features for data processing.

Triple Soft 5600 system with MetabolitePilot software together form a powerful and complete solution for Meta ID studies. The software is designed from the ground up as an easy to use yet powerful tool to efficiently process high resolution, accurate mass IDE data.

As we will discuss, it incorporates features such as structure driven data analysis, unattended batch processing, multiple peak finding strategies and automatic formula prediction.

With the new version 1.5, we have two entirely new functions that were added to help specifically with data interpretation. They are integrated MS/MS fragment interpretation and integrated correlation of metabolites across multiple samples. This allows us to quickly compare metabolism across multiple species and to easily study kinetics across multiple time points for several metabolites at once.

I will begin by giving an overview of the data processing features, then discuss the interpretation functions in detail.

MetabolitePilot can perform automatic unattended batch processing. For each sample, a separate set of processing parameters can be specified, and multiple control samples can also be used.

For example, in the case of an in vivo study, we may want to use dosing of the blank vehicle as one control and the pre-dose sample as another control. Analog data processing is fully integrated into the software, as well, to allow processing of UV or beta ran [sp] data or other types of analog data.

Data processing in MetabolitePilot is structure driven. When building a processing method, the software will use the parent structure to automatically calculate possible cleavage metabolites and to construct a multiple mass defect filtering table for phase one and phase two metabolites.

The mass defect table can also be used not just for data processing, but it can also be exported for acquisition by using the real time multiple mass defect triggered IDA [sp] function of the 5600 system.

Isotope patterns are also calculated automatically and can be used to find and confirm potential metabolites. The software applies multiple parallel peak finding strategies to mine the raw data. This provides both targeted and non-targeted data processing.

Genetic peak finding performs a completely unbiased sample control comparison. We can also search for predicted metabolites from the [unintelligible] table and use mass defect or isotope pattern base of detection.

Since the peak finding strategies are all applied in parallel, very comprehensive data mining can be achieved. For example, if isotope pattern searching is selected for a chlorinated parent compound, it would be very effective in finding the metabolites that are retaining the chlorine. But, since mass defect and genetic peak finding are also used at the same time and applied in parallel, metabolites involving the loss of the chlorine atom will not be missed. Since high resolution IDA MS/MS data is acquired, search can also be performed based on common exact mass product ions or neutral losses and have the software automatically correlate that back to the full scan survey across MS data.

A customizable confirmation score is also calculated for each potential metabolite in order to aid in data review.

MetabolitePilot also has an integrated formula finder, which uses chemical logic and isotope ratios to propose an element of composition for each potential metabolite that is detected.

The elemental composition of the parent can also be used to constrain the number of possible elemental compositions possible for a potential metabolite. For example, if the parent compound contains two nitrogens, a phase one metabolite will not contain more than two nitrogens. Strings and double bonds can also be used to narrow down the possibilities even further.

The built in compound library allows metabolism scientists to store the parent structures and deference MS/MS spectra in this compound library. This eliminates the need to enter them multiple times for different studies involving the same compound.

A comprehensive accurate mass by transformation library consisting of known phase one and phase two biotransformations is included in the software. It is fully customizable, allowing the entry of additional transformations or the modification of existing ones.

I will now discuss how entirely new functions in MetabolitePilot version 1.5 can help with data interpretation in addition to data processing. The software has a new MS/MS fragment interpretation function, which is integrated directly within the software.

There is also a new correlation function for working with multiple samples and automatically correlating metabolist [sp] across those samples in a single result workspace. MS/MS fragment interpretation is fully integrated within the software. It is part of the results view, and the user can easily switch back and forth between the results table and interpretation for a specific metabolite.

In this screen shot, we are showing how the software performs interpretation of the parent MS/MS. As soon as the user switches to interpretation, the parent structure is automatically loaded on the right hand side. And by clicking assign, the software automatically performs a theoretical fragmentation of the molecule and correlates the fragments with the actual high resolution MS/MS spectrum in the acquired data.

And as the user cycles through the fragments table, the proposed fragment is automatically highlighted on the structure.

This slide shows how the function is used with dealkylation metabolite of Haloperidol. The structure is edited directly on the right hand side to do the proposed structure of the metabolite.

The automatic assignment function is then applied to correlate the proposed structure with the actually observed MS/MS spectrum in the data, as shown on the previous slide. This helps the metabolism scientist to quickly determine if the proposed structure is consistent with the MS/MS data. Once the user is set aside for the proposed structure, all the information is saved directly to the results table. The saved structure is now displayed whenever this entry is selected in the results table. In addition, the entire MS/MS fragment interpretation that was performed is also saved in the results table and can be included in the report. By integrating structure elucidation directly within the Metabolite ID software, data processing, interpretation and reporting are streamlined for a faster and more efficient data analysis.

Another new and very innovative feature in MetabolitePilot software version 1.5 is the integrated correlation of multiple samples. The traditional data analysis approach is to process one sample at a time against a control and then perform manual comparisons across samples.

But, with the increased emphasis on quant/qual and on compliance with the Mist [sp] guidelines, it is becoming important to be able to quickly compare metabolist across multiple species in order to spot the disproportionate metabolite and to quickly review time course information across multiple time points. This new multiple sample correlation feature is a major step forward.

In this slide, we demonstrate how metabolist in a time course study can be easily evaluated across multiple time points. The metabolite results table from the five, 15 and 30 minute time points are loaded into the correlation function. And this can include UV or other analog data along with the MS results.

The software automatically correlates the metabolites across the multiple time points and displays the area counts for each sample, as shown here. Also, all MS and MS/MS data is also correlated.

So, when a metabolite is selected in the table, the software automatically overlays the expected ion chromatograms, the MS and the MS/MS data for that metabolite across all the time points, as shown.

The data can also be visualized using automatic plots of the intensity. One or more metabolite can be selected, and a time course plot is automatically generated.

In this plot, we are comparing the time course information of multiple dioxydation [sp] metabolites of Mepyramine across red liver microsome incubation time points. Plots for the MS or the UV data or other analog data can also be generated. Bar charts can also be generated automatically in the software as another type of visual representation of intensity trends across multiple samples. So, if this was an inter-species comparison, this would help in quickly spotting a metabolite that is disproportionate in one species relative to the others.

Summary tables are also generated automatically for the selected metabolites. This allows a quick review of the presence or absence of metabolites across the samples in the study as shown on the bottom portion of the slide and generation of a summary table of area counts for MS or UV, as shown in the upper portion of the slide.

All tables and plots can be included in the report and can also be copied and pasted directly into PowerPoint or Excel for ultimate flexibility.

In conclusion, technology advancements in both hardware and software come together to enable a more powerful and efficient metabolite ID process from data acquisition to processing and now all the way to interpretation. Software features in MetabolitePilot such as the new integrated MS/MS fragment interpretation and multiple sample correlation functions raise the bar in the tools that are available to help the biotransformation scientists streamline their studies.

I would like to acknowledge the product development team for MetabolitePilot, especially Alina Dindyal-Propescu and Hai Ying, and thank our customers at Johnson and Johnson and Bristol Myers Squibb for their extremely valuable input and feature suggestions during the development process and also for doing the beta testing.

Thank you very much for your attention.