Join our next WEBINAR:

Transfer of spectroscopic calibrations:
strategies and best practices!

In a constantly evolving industrial context, the transfer of spectroscopic calibrations represents a real challenge for laboratories and production sites.

Indeed, each campaign, each reference measurement, each spectrum represents a significant cost and a substantial investment of time.

Thanks to inter-instrument data transfer, preserve the integrity of your databases: spectra, reference measurements, and past campaigns.

• Don’t waste years of work!
• Avoid having to remeasure all your samples on your new equipment.
• Reduce model development and maintenance time.
• Ensure the continuity of your analyses and the sustainability of your models to guarantee the reliability of your results.

Whether your goal is to upgrade your instrument portfolio, adapt your methods to new environments (field, online), or scale up to industrial scale, data preservation and calibration reliability are essential to ensuring the quality of your analyses.

Join us on June 11, 2026, at 4.00 pm (CEST) for a webinar dedicated to spectral calibration transfer methods.

REGISTER
HERE

The calibration transfer shall not be taken lightly. During this webinar, we will show you:

• the workflow to follow, depending on the available standardization samples
• how to evaluate the performance of a calibration transfer
• the pitfalls to avoid: overfitting, instrumental biases, lack of data representativeness…
• the best practices for performing an efficient calibration transfer

This webinar will be illustrated by a practical case study demonstrating the performance of different transfer methods on spectral data.

The post June 11, 2026 – “Transfer of spectral calibrations: strategies and best practices!” appeared first on Ondalys.

We invite you to our next webinar on Machine Learning methods and the interpretation of ML models developed on spectral databases on March 10, 2026 at 4 p.m. (CET).

Would you like to learn more about how to make your
Machine Learning models built on spectroscopic data more reliable? 

Webinar program:

• Why use Machine Learning (ML) models?
• The interest of ML models interpretability
• Overview of ML methods
• Principles and applications of kernel methods and neural networks
• Overview of interpretability methods of ML models
• Principles and applications of interpretability methods (SHAP, etc.) to spectroscopic data
• Q&A session

Dr. Astrid Maléchaux, Data Scientist at Ondalys for several years and expert in Machine Learning and model interpretability, will present this webinar. With her, you will discover how to avoid the black box effect of some Machine Learning models on spectral data and ensure the reliability of these models.

Experts in spectral calibration and in developing Machine Learning methods adapted to spectral data (NIR, MIR, Raman, …), Ondalys team supports you in analyzing your spectroscopic data and developing spectral calibrations.

The post March 10, 2026 – Webinar Optimization et interpretation of ML models appeared first on Ondalys.

This year again, Ondalys is sponsoring the 25th Chimiometrie Conference. Come and discuss with our team at this event.

2 occasions to discuss Chemometrics with our team !

The 2026 Chimiometrie congress is organized by the École Nationale Supérieure des Industries Chimiques (ENSIC) on the Grandville campus in downtown Nancy, France.

This annual meeting of industrial and academic scientifics interested in Chemometrics and data analytics will give our team the opportunity to share with you our expertise on the topics covered during the oral presentations:

• Spectroscopic data analysis,
• Hyperspectral imaging analysis,
• Design of Experiments – DoE
• OMICS data analysis (LC-MS, GC-MS, GCxGC, etc.),
• Process monitoring and optimisation
• and many other Chemometrics methods

More information soon

See more details on the Chimiometrie 2026 conference website.

The 2026 edition of the conference will be held in Nancy on the Grandville campus of the École Nationale Supérieure des Industries Chimiques (ENSIC).

Ondalys offers data analysis services and training and distributes some of the most widely used and recognized Chemometrics software, such as PLS_Toolbox® and SOLO® for Chemometrics, MIA Toolbox® for hyperspectral image analysis from EigenVector Research Inc., and Design-Expert® software from Stat-Ease for Design of Experiments.

The Chemometrics Art

At the heart of data analysis lies Chemometrics, a discipline that focuses on the complex relationships between different variables. In a context where data is vast and often intertwined, Chemometrics provides a powerful way to explore interactions between variables and discover those that have a significant impact on outcomes.

Rather than examining variables individually, Chemometrics seeks to capture the dynamics of interactions. This approach offers a holistic view of the data, uncovering hidden patterns and complex correlations.

By analyzing how variables interact with each other, Chemometrics makes it possible to identify those with the greatest influence on final results. These key variables may directly affect product quality, process performance, or other critical industrial metrics.

Chemometrics offers sophisticated techniques such as Principal Component Analysis (PCA), Partial Least Squares regression (PLS), and other advanced methods to unravel complex interactions. These approaches provide deeper insights into relationships between variables, thereby supporting better decisions and actions.

By integrating the art of Chemometrics into your industrial data analysis, raw datasets are transformed into actionable knowledge. By recognizing hidden influences and dependencies among variables, you can make informed decisions that drive your business toward efficiency and competitiveness.

The post February 17-19, 2025 : Chimiometrie 2026 Congress in Nancy, France appeared first on Ondalys.

New scientific article!

Monitoring of monoclonal antibody critical quality attributes and free amino acids in CHO bioprocesses using Raman spectroscopy and DUPLEX-based PLS modeling. 

Published in Microchemical Journal in late September, this article presents work carried out on the online monitoring of Critical Quality Attributes (CQAs) of monoclonal antibodies (mAbs) and free amino acids in CHO cultures using Raman spectroscopy.

The PLS (Partial Least Square) modeling made it possible the prediction of the final quality criteria, the following CQAs: :

• Aggregates – multimeric monoclonal antibodies which affect stability and immunogenicity.
• Fragments – degradation or cleavage products of the mAb, impacting purity and efficiency.
• Glycoforms – variations in glycosylation profile, influencing pharmacokinetics and bioactivity.
• Charged variants – isoforms resulting from post-translational modifications (e.g., deamidation, amidation, C-terminal lysine), which affect stability and therapeutic performance.

🧪 Moreover, the developed models have also made it possible to predict free amino acids in the culture medium (dynamic nutritional profile), which allows simultaneous monitoring of cellular metabolism and product quality.

Morandise Rubini conducted this research as part of the CLIMBIN project with various researchers from the University of Tours, Indatech-Chauvin Arnoux Group, Servier Technology, and Ondalys. Jordane Poulain and Sylvie Roussel, co-authors of this article, contributed their expertise in Machine Learning.

Morandise Rubini, Anaïs Berger, Thomas Saillard, Fabrice Cantais, Martin Soucé, Jordane Poulain, Sylvie Roussel, Julien Louet, Fabien Chauchard-Rios, Sylvain Arnould, Igor Chourpa. (2025). Monitoring of monoclonal antibody critical quality attributes and free amino acids in CHO bioprocesses using Raman spectroscopy and DUPLEX-based PLS modeling. Microchemical Journal 218 (2025) 115583

The post Sept 2025 – Publication : Online monitoring of mAbs’ CQAs using Raman. appeared first on Ondalys.

Discover the results of this study

The Colloquium Chemiometricum Mediterraneum is a scientific event that discusses the latest innovations in Chemometrics, with presentations by French, Spanish, Portuguese, and Italian chemometricians in their respective own languages. The CCM is a place to intiate new international collaborations within the chemometrics community.

The post Sept. 9-12, 2025 – Colloquium Chemiometricum Mediterraneum appeared first on Ondalys.

We are happy to announce our next webinaire jointly given with our partner, the spectrometer manufacturer Metrohm USA.

In this webinar, you will discover how to implement Online Spectroscopy techniques such as Raman and NIR to :

• improve process control
• ensure consistent product quality 
• reduce waste and consumables

Elena Hagemann, Product Manager Process Spectroscopy at Metrohm USA, and Sylvie Roussel, Ondalys CEO, will present practical guidances for identifying suitable applications, defining technical requirements for robust interface design, developing reliable prediction models, and integrating spectroscopy into existing control systems, all to support confident decision-making and enhance process performance.

This webinar will focus on these Key points:

• Implementation Strategy: Learn how to assess feasibility, avoid common pitfalls, and plan for long-term success with online spectroscopy.
• System Design & Integration: Explore best practices for sampling interface design and seamless integration with DCS/PLC systems.
• Calibration & Reliability: Understand how to develop robust calibration models and maintain analyzer accuracy through monitoring, diagnostics, and lifecycle support.

Expert in spectral data analysis (NIR, MIR, Raman, UV), the Ondalys team supports you for all your issues of spectroscopic data processing and spectral calibration development.

The post July 23, 2025 – Webinar: Implementing Online Spectroscopy appeared first on Ondalys.

Organized by SISNIR – Società Italiana di Spettrosocopia NIR, the 22^nd International Conference on Near Infrared Spectroscopy – NIR2025, will be held from June 8^th toJune 12^th 2025 in Roma, Italy.

Expert in Chemometrics and Machine Learning methods applied to spectral data, the ondalys team will attend this conference to discuss about spectroscopic data analysis and will present 2 communications on the following themes :

Machine Learning interpretability methods applied to calibration models developed on NIR spectroscopic data

In the past decades, Machine Learning (ML) models have become more and more complex, leading to improvements in their predictive performance. However, these models can often be described as “black boxes”, in the sense that it is very difficult to explain how results are obtained by a model from the input data. As complex Machine Learning models are increasingly used to make decisions, for instance in industrial or medical applications, there is a growing need to improve their interpretability in order to have greater confidence in their results, providing the so-called Explainable AI (Artificial Intelligence).

This study is focused on the interpretability of Machine Learning models after model calibration on near-infrared spectroscopic data, a.k.a. the “post-hoc explanation” of ML models.

Learn more about the advanced Machine Learning methods

Machine Learning methods for sugar quantification in grapes based on NIR Hyperspectral Imaging

Hyperspectral imaging [1] has applications in many fields, including agriculture, environment, medicine and industry. It can be used to classify objects according to their composition, or to quantify compounds present on the surface and show their spatial distribution. This last case is often challenging as the reference is not known for each pixel. Image processing is even more important in this case to extract the Region of Interest (ROI) to build the model but also to be able to apply it on new images.

This scientific study presents the methodology for analyzing hyperspectral images for grape bunch maturity prediction directly in vineyards. In this example, the parameter to be predicted is the sugar content (in °Brix) on each whole bunch of grapes, providing an average measurement of sugar content per image

Learn more about how and why analysizing hyperspectral images – HSI

NIR2025 will highlight all aspects of near-infrared spectroscopy, spanning from methodological and technological advances in NIR spectroscopy (near-infrared imaging and miniaturized spectrometers) to applications in various fields (agri-food, pharmaceutical, chemical, biotech, etc.) through the analysis of spectroscopic data and hyperspectral images.

Every two years, in fact, the International Council for Near Infrared Spectroscopy (ICNIRS – https://icnirs.org/) organizes the main international conference dedicated to NIR spectroscopy.
Each edition permits to to catch up on all the NIR news (theory, applications, innovative instrumental solutions).

The post June 8-12, 2025 – NIR2025 international congress in Roma, Italy appeared first on Ondalys.

This work aimed at developing a real-time cell culture monitoring method to optimize the culture conditions of CHO (Chinese hamster ovary cells) for the production of monoclonal antibodies; and to guarantee a constant quality of the main metabolic parameters and the IgG (immunoglobulins) yield for biopharmaceutical production.

Raman spectroscopy, combined with Machine Learning methods for spectral processing, has established itself as a robust process analytical technology (PAT) tool thanks to its real-time, in-situ, and non-invasive measurement capabilities in bioreactors. This study evaluates online Raman spectroscopy and chemometric and machine learning methods for in-process prediction, with a view to monitoring the following parameters of interest: nutrients, metabolites, antibody titers, and cell density.

Following the development of regression models using chemometric and machine learning methods by varying cell culture conditions, this study confirms the potential of Raman spectroscopy for in-situ and real-time monitoring of bioprocesses, particularly for CHO cell cultures and monoclonal antibody production, without manual sampling. Chemometric analysis improves the accuracy and robustness of the models and enables the monitoring and even automated control of bioreactors. Raman data could enable continuous regulation of critical nutrients such as glucose, thus ensuring control of Critical Process Parameters (CPPs) during biopharmaceutical production.

This work was conducted within the framework of the CLIMBIN collaborative R&D project, in collaboration with various researchers from the University of Tours (Dr. Morandise Rubini, Prof. Igor Chourpa), the LRGP laboratory in Nancy, the Servier Group (Anaïs Berger, Thomas Saillard, Sylvain Arnould, Muriel Vergès), the spectroscopy equipment manufacturer INDATECH (Julien Louet, Dr. Fabien Chauchard), and Ondalys (Julien Boyer, Jordane Poulain, Dr. Sylvie Roussel).

For this article, Julien Boyer, Jordane Poulain and Dr. Sylvie Roussel from Ondalys contributed their expertise in Chemometrics and Machine Learning. They brought their experience for data processing using online Raman spectroscopy measurements for the bioproduction of mAbs from CHO cells under industrial conditions, in Servier laboratories, alongside the University of Tours.

M. Rubini, J. Boyer, J. Poulain, A. Berger, T. Saillard, J. Louet, M. Soucé, S. Roussel, S. Arnould, M. Vergès, F. Chauchard-Rios & I. Chourpa. Monitoring of Nutrients, Metabolites, IgG Titer, and Cell Densities in 10 L Bioreactors Using Raman Spectroscopy and PLS Regression Models. Pharmaceutics 2025, Volume 17, Issue 4, 473, Avril 2025

The post April 2025 – Scientific article: Cell culture monitoring using Raman and Machine Learning appeared first on Ondalys.

👉 ALL you need to know about France Bioproduction Congress 2025

With extensive experience in supporting biotechnology and biopharmaceutical manufacturers, the Ondalys team offers its expertise to address issues related to monitoring and optimizing cell bioproduction (yeasts, CHO, antibodies, etc.).

Watch Sylvie Roussel interview
during the 2023 France Bioproduction Congress

To learn more about Machine Learning and Data Analytics for the Biopharma and Biotech industries,
check out our dedicated pages.

Polepharma is the 1st pharmaceutical cluster and aims to structure the French (bio)pharmaceutical industrial sector by supporting its transformation towards the Industry of the Future.

The mission that Polepharma has set itself revolves around 3 objectives:

• Make France the best place in the world to develop & produce (bio)medicines.
• Manufacture quality treatments in France in the best conditions and secure public health, contribute to employment and local economies.
• Promote industrial cooperation between territories and laboratories, suppliers, schools, collectives, experts, talents, etc.

The post March 2025 – French Bioproduction Congress Polepharma – CFB 2025 appeared first on Ondalys.

The post The Ondalys team sends you its best wishes for 2025 ! appeared first on Ondalys.