What’s New in v2019 of ACD/Spectrus and ACD/Percepta
No Man Nor Organization is an Island | A Meditation on the Importance of Partnerships for Scientific Innovation

Can Bias be Totally Removed from Structure Verification?

Yalda Mobarhan, Marketing Communications Specialist

Reliable structure verification of candidate molecules is at the very core of any successful chemical discovery or development program. NMR provides unprecedented levels of information for unraveling the actual structure of compounds, and is an essential tool for chemists to confirm that they made what they intended to make.

Image001

In many labs in addition to synthesis, synthetic/medicinal chemists must also perform the majority of routine structure confirmation, data analysis, and interpretation, with minimal supervision by an expert spectroscopist.

Unfortunately, no chemical process is 100% error-free; uncharacterized raw materials, unexpected synthetic routes, unaccounted for structural isomers, etc., may all lead to unexpected final products. For example, in the Bosutinib case where structural isomers instead of the genuine molecule were produced and unwittingly administered to humans. Similar issues can be avoided as long as enough attention is paid to thorough interpretation of the analytical data, specifically NMR, to quickly and accurately verify the structures. This is not always an easy task and may be entrusted to automated systems for execution, since the chemist’s expected outcomes may influence the result (bias).

ACD/Labs’ Automated Structure Verification (ASV) tool was developed in 2003 to deliver a robust and high performance verification system to pharmaceutical and chemical organizations. ASV has evolved and improved significantly since its launch. The tool is based on a combination of advanced NMR processing and prediction algorithms. In its simplest form, the software predicts the spectral properties of a chemical structure proposed by a “human expert” during verification, and matches it against the experimental dataset. Although constitutional isomers represent alternative structures that could also match the spectral data, these isomers are neither evaluated nor ruled out. This method uses the predicted chemical shifts, multiplicity, and intensity information to evaluate the “proposed structure” fit with the experimental spectrum.

In 2014, the Combined and Concurrent Verification (CCV) method was developed by ACD/Labs to reduce false positive rates using multi-step verification. Specifically, CCV only evaluates a selected number of alternatives to the structure proposed by the user. These structures are generated by moving molecular substituents along a ring or a chain, thus other structures consistent with the NMR data are inherently ignored. Therefore, the initial bias from the chemist may affect the system by producing a false positive result.

The next step in ASV development was to produce a tool that eliminates user bias completely. Several attempts were made to remove this bias by requiring multiple 2D NMR datasets which, unfortunately, resulted in longer and more costly analytical procedures.

Image002

ACD/Labs successfully removed user bias from the structure verification workflow in their software by releasing the Unbiased Verification (UBV) tool in 2017. UBV is based on the fastest structure generation engine in the market (implemented for more than two decades in ACD/Structure Elucidator), coupled with the best ranking system used in Computer Assisted Structure Elucidation (CASE). In this workflow, the molecular formula is the only given parameter used from the proposed structure by extraction directly from mass spectral data of the sample in question. The molecular formula is used by the software to generate all plausible structures. The entire process is seamless to setup and will produce a quality ranked list of structures that best match the spectra, thus totally eliminating user intervention. UBV has been shown to outperform previous ASV techniques in complex cases where the “best structure” is not essentially the “proposed structure” chosen by the expert.

UBV represents the most efficient workflow for rapidly recognizing correct structures, while at the same time flagging structures that are chemically relevant, yet incorrect, for further analysis by the "human expert". Furthermore, the improved productivity, reliability and reduced chemist supervision involved enables a greater return on analytical investment. ASV implementation is particularly beneficial to synthetic chemists, as more focus can be devoted to actual synthesis, and less time and effort spent on characterizing molecules. In turn, researchers and executives can ensure that target compounds are produced in both sufficient quality and quantity.

In general, the advantages of tools such as unbiased structure verification are relevant to us all. Faster, cheaper and more reliable structure characterization can help to reduce the risk and expense associated with research on novel molecular entities and hopefully lead to faster development of new chemicals including pharmaceuticals.

The UBV tool is included in ACD/NMR Workbook Suite—the ultimate tool for processing and analysis of NMR data. For more info visit www.acdlabs.com/NMRWorkbookSuite

 

Comments

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)