2025 Structure Elucidation & Verification Report
In October 2024, ACD/Labs hosted our third annual virtual symposium “Unveiling Spectral Secrets—Structure Elucidation & Verification Virtual Symposium”. In the past, these events have featured presentations from scientists at leading pharmaceutical and chemical organizations including Pfizer, Merck, Medtronic, Bristol Myers Squibb, and MilliporeSigma (known as Merck KGaA outside of the US and Canada).
This year revisited the structure verification and elucidation theme from the inaugural 2022 event. We compared how opinions and outlooks have evolved over the last few years among a similarly composed audience of stakeholders and scientists.
The 2024 symposium featured six presentations from scientists at top R&D organizations across chemical industries:
Emine Sager
Analytical Scientist, Novartis
With over 20 years of experience supporting pharmaceutical research, Emine plays a key role in the Analytical NMR laboratory at Novartis.
Christine Fisher (O’Donnell)
Chemist, U.S Food and Drug Administration (FDA)
Christine’s current research focuses on developing non-targeted analysis (NTA) methods using high-resolution mass spectrometry for food safety applications.
Kristof Cank
Analytical Scientist, Sutro Biopharma
Kristof is charged with identifying impurities and evaluating the purity, potency, and stability of compounds in the development of next generation cancer therapeutics.
Richard Lewis
Principal Scientist, AstraZeneca
Richard has had a long and distinguished career in NMR. He currently leads the NMR Center of excellence.
Jennifer Field
Technical Laboratory Supervisor, Shimadzu
Jen brings a wealth of expertise in chromatography and method development to evaluating innovative instrumentation and methods development for novel applications.
Jens Richards
Analytical Scientist, Corteva Agriscience
Jens leverages his expertise to advance chromatography development and establish computer-assisted method development.
Efficiency Gains from Modeling and Automation
Our group of experts highlighted a trend towards automating workflows to handle increasing sample volumes and complexity. They showcased how automation and high-throughput analysis are being applied across the diverse sectors of the analytical chemistry field they represent for experiment planning, data analysis, and reporting.
Efficient Method Development & Optimization with Retention Modeling
A common first step in structure elucidation or verification is separating the compound from a mixture. Retention time and mass spectral data from hyphenated chromatography techniques also aid in identifying and verifying known unknowns. This year, 75% of respondents indicated using chromatography for structure elucidation when we added them as explicit options in our poll.
Jen from Shimadzu emphasized retention modeling as an important step to minimize resource usage while improving method robustness and sustainability. She demonstrated how in silico retention modeling can significantly increase throughput and efficiency of chromatographic method development noting that in silico retention modeling could provide the information equivalent to 600 experiments in just 4 experimental runs.
At Corteva, Jens’ team recently took a new approach to chromatographic method development, which included the use of AutoChrom to support more efficient method development. By the time of the presentation, they were only slightly behind their ambitious target of developing 75 methods for the year through their software-assisted chromatographic method development workflow (SPEEDChrom). This represents a significant efficiency gain and allows new members to quickly onboard and gain valuable experience.
Meanwhile, Christine explained how they find data processing to be the most time-consuming step in the FDA’s non-targeted analysis (NTA) and suspect screening analysis (SSA) workflow. They use retention prediction to eliminate false positive results from NTA. Fewer candidates moving on to the next portion of the workflow means less data to process.
Automated Structure Verification (ASV) by NMR
Richard and Emine focused on the use of NMR data for structure elucidation & verification. They highlighted automated structure verification (ASV) as an important tool for both NMR experts and synthetic/medicinal chemists to handle large data volumes and simplify complex molecular analysis.
“We always advise our chemists to use ASV as a supporting tool, which significantly saves time and improves efficiency in the assignment process.” – Emine Sager, Novartis
Kristof also reported that ASV is among the software supports employed in their antibody-drug conjugate development process. They also use other Spectrus applications to automate processing of LC/MS and HPLC data and report generation, enabling high-throughput analysis.
The audience showed strong interest in automating data analysis for structure elucidation and verification. As in 2022, hyphenated chromatography techniques remain the top priority for automation, though interest in automating NMR and MS data has grown.
Despite strong interest, adoption of automated structure elucidation and verification has seen little growth. In fact, the data suggests a decline, with fewer respondents reporting existing automation and only modest increases in those exploring or testing such solutions.
Improving Data Quality & Accuracy in Structure Elucidation & Verification
As we saw in 2022, automated solutions like ASV offer significant efficiency gains. Improving accuracy remains a key concern for users and stakeholders to continue enhancing efficiency.
Eliminating the Risk of Data Transcription Errors
Several speakers highlighted the importance of high-quality empirical data for analysis, modeling, and prediction to optimize accuracy and efficiency, especially in novel chemical spaces. They also noted that all-in-one software, like Spectrus, streamlines data analysis and management from multiple techniques, aiding knowledge transfer and reducing human errors, such as in transcription.
Jens remarks that AutoChrom helps them organize the massive amount of data they generate, which is “really important in our current state of technology to iterate and deliver solutions faster.” They use AutoChrom to both design their experiments and control the instrument, which eliminates human error from the outset. Finally, he adds:
“[W]e can model conditions based on empirical data, and this is really, really important for us. We need to be able to get real data on our analogs and then make decisions from that data. So [AutoChrom] not only enables us to collect more data, but makes the data that we’re collecting more powerful.” Jens Richards, Corteva Agriscience
In her discussion of the potential pitfalls in retention modelling, Jennifer also describes how all-in-one software, such as the Spectrus applications, can help minimize the risk of human error.
“Error is often associated with typos or transcription errors. This can occur when the data processing is not connected to the modeling software. However, if data is processed with ACD/Labs Spectrus Processor, it can be transferred seamlessly into LC Simulator, minimizing this risk.” Jennifer Field, Shimadzu
Enhance Accuracy by Tailoring Analysis to Your Samples
Several presentations addressed the challenges of analyzing and verifying complex molecules, highlighting the need for tailored approaches to ensure accurate high-throughput workflows. Jen discussed how Shimadzu adapts models with different mathematical relationships for biomolecules versus small molecules.
On the empirical side, Emine shared how Novartis’ ASV workflow first records a proton spectrum, which is analyzed using an in-house script to optimize subsequent 13C and 2D NMR experiments. She provided a glimpse into their results with a one-day summary from their open-access system. All but 4% of samples had experimental data that was applicable to their ASV system. The structure was accurately verified (true positive) or rejected (true negative) for 40% and 15% of samples respectively. Another 39% of samples were wrongly identified as having the incorrect structure proposed (false negative) while only 2% were wrongly verified as the correct structure (false positive).
Overall, they concluded that while there is room for improvement, the ASV system is sufficiently accurate to provide efficiency benefits without significantly increasing risk. (The two false positive results were caused by issues with the sample and selection of experiments).
Open-Access ASV Results at Novartis
- 116 samples
- Various experiments—from simple 1H to comprehensive 2D sets
- Wide variety of molecules
- MW 126-1168 Da
- Heteroatoms: F, P
- Different heterocycles
Get a Clearer Picture with Integration of Multiple Analytical Techniques
Finally, many presentations highlighted the combination of different analytical methods employed to improve structure elucidation and verification.
Kristof and the small molecule analytical chemistry team at Sutro Biopharma use a mix of analytical methods—LC/MS, NMR, supercritical fluid chromatography (SFC), and 2D-LC—for purity analysis, impurity identification, and structural elucidation of linkers, warheads, leads, and side products in their antibody-drug conjugate (ADC) development process. They rely on ACD/Labs software to process all this data within a single platform, enabling more efficient analysis, especially for closely related compounds or impurities. Kristof also highlighted using NMR Predictors alongside experimental data to enhance confidence in structural analysis, particularly for compounds with only small differences in functional groups.
“One of the advantages of ACD/Labs [software] is that all types of data can be processed here: high resolution, TOF data, single quad MS, HPLCs, NMR, flash chromatograms, all processed in the same software and linked together.” Kristof Cank, Sutro Biopharma
Richard is also strongly in favour of combining data from multiple analytical techniques for efficient and accurate structure verification. He detailed the extensive studies that AstraZeneca have undertaken to investigate the accuracy of combining NMR, MS, and IR spectroscopy data for ASV. He sums it up by saying:
“I think the future is likely to be a different mix of different approaches. So not just one bit of software, one bit of data, but putting lots of different software and data together to get an answer.” Richard Lewis, AstraZeneca
AI/ML Integration for Structure Verification
The need for high-quality data is underscored by the trend towards using machine learning (ML) and artificial intelligence (AI) for structure elucidation and verification. While not explicitly mentioned by everyone, the use of such technologies is already pervasive, as ML algorithms are foundational to the ACD/Labs’ modeling and prediction software discussed previously.
However, when it comes to relying on AI and ML to make decisions in structure elucidation and verification workflows, Richard’s presentation of AstraZeneca’s work to improve ASV is a good example of work being done as a foundation towards AI-powered structure elucidation.
He explains that in their workflow, they use HRMS data, which allows them to know the molecular formula with a high degree of certainty prior to analyzing the NMR data. Therefore, such a system should be able to use the molecular formula along with minimal NMR data (to keep the workflow high throughput), and verify structures with a high degree of accuracy, in order to distinguish isomeric reaction products. He adds that they can tolerate a system that is wrong occasionally, however, it should “know when it doesn’t know” so that a scientist can then step in to provide more data or perform the interpretation manually. Scientists at AZ continue to develop this technology to alleviate the burden of routine SE&V.
Key Trends & Insights for Modern Structure Characterization
The Structure Elucidation and Verification Virtual Symposium highlighted key advancements in the field, emphasizing the growing role of automation, data integration, and multi-technique approaches. While adoption of automated tools like retention modeling and ASV has faced challenges, the potential for improved efficiency and accuracy remains clear. The experts stressed the importance of high-quality data and tailored workflows to minimize errors and optimize results. Looking ahead, collaboration and the integration of emerging technologies, such as AI, promise to drive further innovation, empowering the analytical community to meet evolving challenges and accelerate discovery.