̽��ֱ�� of Cambridge - Alpha Lee

Accelerating how new drugs are made with machine learning

sc604 — Mon, 15 Jan 2024 10:05:29 +0000

Predicting how molecules will react is vital for the discovery and manufacture of new pharmaceuticals, but historically this has been a trial-and-error process, and the reactions often fail. To predict how molecules will react, chemists usually simulate electrons and atoms in simplified models, a process that is computationally expensive and often inaccurate.

Now, researchers from the ̽��ֱ�� of Cambridge have developed a data-driven approach, inspired by genomics, where automated experiments are combined with machine learning to understand chemical reactivity, greatly speeding up the process. They’ve called their approach, which was validated on a dataset of more than 39,000 pharmaceutically relevant reactions, the chemical ‘reactome’.

Their results, reported in the journal Nature Chemistry, are the product of a collaboration between Cambridge and Pfizer.

“ ̽��ֱ��reactome could change the way we think about organic chemistry,” said Dr Emma King-Smith from Cambridge’s Cavendish Laboratory, the paper’s first author. “A deeper understanding of the chemistry could enable us to make pharmaceuticals and so many other useful products much faster. But more fundamentally, the understanding we hope to generate will be beneficial to anyone who works with molecules.”

̽��ֱ��reactome approach picks out relevant correlations between reactants, reagents, and performance of the reaction from the data, and points out gaps in the data itself. ̽��ֱ��data is generated from very fast, or high throughput, automated experiments.

“High throughput chemistry has been a game-changer, but we believed there was a way to uncover a deeper understanding of chemical reactions than what can be observed from the initial results of a high throughput experiment,” said King-Smith.

“Our approach uncovers the hidden relationships between reaction components and outcomes,” said Dr Alpha Lee, who led the research. “ ̽��ֱ��dataset we trained the model on is massive – it will help bring the process of chemical discovery from trial-and-error to the age of big data.”

In a related paper, published in Nature Communications, the team developed a machine learning approach that enables chemists to introduce precise transformations to pre-specified regions of a molecule, enabling faster drug design.

̽��ֱ��approach allows chemists to tweak complex molecules – like a last-minute design change – without having to make them from scratch. Making a molecule in the lab is typically a multi-step process, like building a house. If chemists want to vary the core of a molecule, the conventional way is to rebuild the molecule, like knocking the house down and rebuilding from scratch. However, core variations are important to medicine design.

A class of reactions, known as late-stage functionalisation reactions, attempts to directly introduce chemical transformations to the core, avoiding the need to start from scratch. However, it is challenging to make late-stage functionalisation selective and controlled – there are typically many regions of the molecules that can react, and it is difficult to predict the outcome.

“Late-stage functionalisations can yield unpredictable results and current methods of modelling, including our own expert intuition, isn't perfect,” said King-Smith. “A more predictive model would give us the opportunity for better screening.”

̽��ֱ��researchers developed a machine learning model that predicts where a molecule would react, and how the site of reaction vary as a function of different reaction conditions. This enables chemists to find ways to precisely tweak the core of a molecule.

“We trained the model on a large body of spectroscopic data – effectively teaching the model general chemistry – before fine-tuning it to predict these intricate transformations,” said King-Smith. This approach allowed the team to overcome the limitation of low data: there are relatively few late-stage functionalisation reactions reported in the scientific literature. ̽��ֱ��team experimentally validated the model on a diverse set of drug-like molecules and was able to accurately predict the sites of reactivity under different conditions.

“ ̽��ֱ��application of machine learning to chemistry is often throttled by the problem that the amount of data is small compared to the vastness of chemical space,” said Lee. “Our approach – designing models that learn from large datasets that are similar but not the same as the problem we are trying to solve – resolves this fundamental low-data challenge and could unlock advances beyond late-stage functionalisation.”

̽��ֱ��research was supported in part by Pfizer and the Royal Society.

References:
Emma King-Smith et al. ‘Predictive Minisci Late Stage Functionalization with Transfer Learning.’ Nature Communications (2023). DOI: 10.1038/s41467-023-42145-1

Emma King-Smith et al. ‘Probing the Chemical "Reactome" with High Throughput Experimentation Data.’ Nature Chemistry (2023). DOI: 10.1038/s41557-023-01393-w

Researchers have developed a platform that combines automated experiments with AI to predict how chemicals will react with one another, which could accelerate the design process for new drugs.

A deeper understanding of the chemistry could enable us to make pharmaceuticals and so many other useful products much faster.

Emma King-Smith

BlackJack3D via Getty Images

Digital Molecular Structure Concept

̽��ֱ��text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Machine learning algorithm predicts how to get the most out of electric vehicle batteries

sc604 — Tue, 23 Aug 2022 09:01:34 +0000

̽��ֱ��researchers, from the ̽��ֱ�� of Cambridge, say their algorithm could help drivers, manufacturers and businesses get the most out of the batteries that power electric vehicles by suggesting routes and driving patterns that minimise battery degradation and charging times.

̽��ֱ��team developed a non-invasive way to probe batteries and get a holistic view of battery health. These results were then fed into a machine learning algorithm that can predict how different driving patterns will affect the future health of the battery.

If developed commercially, the algorithm could be used to recommend routes that get drivers from point to point in the shortest time without degrading the battery, for example, or recommend the fastest way to charge the battery without causing it to degrade. ̽��ֱ��results are reported in the journal Nature Communications.

̽��ֱ��health of a battery, whether it’s in a smartphone or a car, is far more complex than a single number on a screen. “Battery health, like human health, is a multi-dimensional thing, and it can degrade in lots of different ways,” said first author Penelope Jones, from Cambridge’s Cavendish Laboratory. “Most methods of monitoring battery health assume that a battery is always used in the same way. But that’s not how we use batteries in real life. If I’m streaming a TV show on my phone, it’s going to run down the battery a whole lot faster than if I’m using it for messaging. It’s the same with electric cars – how you drive will affect how the battery degrades.”

“Most of us will replace our phones well before the battery degrades to the point that it’s unusable, but for cars, the batteries need to last for five, ten years or more,” said Dr Alpha Lee, who led the research. “Battery capacity can change drastically over that time, so we wanted to come up with a better way of checking battery health.”

̽��ֱ��researchers developed a non-invasive probe that sends high-dimensional electrical pulses into a battery and measures the response, providing a series of ‘biomarkers’ of battery health. This method is gentle on the battery and doesn’t cause it to degrade any further.

̽��ֱ��electrical signals from the battery were converted into a description of the battery’s state, which was fed into a machine learning algorithm. ̽��ֱ��algorithm was able to predict how the battery would respond in the next charge-discharge cycle, depending on how quickly the battery was charged and how fast the car would be going the next time it was on the road. Tests with 88 commercial batteries showed that the algorithm did not require any information about previous usage of the battery to make an accurate prediction.

̽��ֱ��experiment focused on lithium cobalt oxide (LCO) cells, which are widely used in rechargeable batteries, but the method is generalisable across the different types of battery chemistries used in electric vehicles today.

“This method could unlock value in so many parts of the supply chain, whether you’re a manufacturer, an end user, or a recycler, because it allows us to capture the health of the battery beyond a single number, and because it’s predictive,” said Lee. “It could reduce the time it takes to develop new types of batteries, because we’ll be able to predict how they will degrade under different operating conditions.”

̽��ֱ��researchers say that in addition to manufacturers and drivers, their method could be useful for businesses that operate large fleets of electric vehicles, such as logistics companies. “ ̽��ֱ��framework we’ve developed could help companies optimise how they use their vehicles to improve the overall battery life of the fleet,” said Lee. “There’s so much potential with a framework like this.”

“It’s been such an exciting framework to build because it could solve so many of the challenges in the battery field today,” said Jones. “It’s a great time to be involved in the field of battery research, which is so important in helping address climate change by transitioning away from fossil fuels.”

̽��ֱ��researchers are now working with battery manufacturers to accelerate the development of safer, longer-lasting next-generation batteries. They are also exploring how their framework could be used to develop optimal fast charging protocols to reduce electric vehicle charging times without causing degradation.

̽��ֱ��research was supported by the Winton Programme for the Physics of Sustainability, the Ernest Oppenheimer Fund, ̽��ֱ��Alan Turing Institute and the Royal Society.

Reference:
Penelope K Jones, Ulrich Stimming & Alpha A Lee. ‘Impedance-based forecasting of lithium-ion battery performance amid uneven usage.’ Nature Communications (2022). DOI: 10.1038/s41467-022-32422-w

For more information on energy-related research in Cambridge, please visit Energy IRC, which brings together Cambridge’s research knowledge and expertise, in collaboration with global partners, to create solutions for a sustainable and resilient energy landscape for generations to come.

Researchers have developed a machine learning algorithm that could help reduce charging times and prolong battery life in electric vehicles by predicting how different driving patterns affect battery performance, improving safety and reliability.

This method could unlock value in so many parts of the supply chain, whether you’re a manufacturer, an end user, or a recycler, because it allows us to capture the health of the battery beyond a single number

Alpha Lee

Monty Rakusen via Getty Images

People charging their electric cars at charging station in York

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

AI tackles the challenge of materials structure prediction

Anonymous — Wed, 27 Jul 2022 17:46:49 +0000

̽��ֱ��researchers, from Cambridge and Linkoping Universities, have designed a way to predict the structure of materials given its constitutive elements. ̽��ֱ��results are reported in the journal Science Advances.

̽��ֱ��arrangement of atoms in a material determines its properties. ̽��ֱ��ability to predict this arrangement computationally for different combinations of elements, without having to make the material in the lab, would enable researchers to quickly design and improve materials. This paves the way for advances such as better batteries and photovoltaics.

However, there are many ways that atoms can ‘pack’ into a material: some packings are stable, others are not. Determining the stability of a packing is computationally intensive, and calculating every possible arrangement of atoms to find the best one is not practical. This is a significant bottleneck in materials science.

“This materials structure prediction challenge is similar to the protein folding problem in biology,” said Dr Alpha Lee from Cambridge’s Cavendish Laboratory, who co-led the research. “There are many possible structures that a material can ‘fold’ into. Except the materials science problem is perhaps even more challenging than biology because it considers a much broader set of elements.”

Lee and his colleagues developed a method based on machine learning that successfully tackles this challenge. They developed a new way to describe materials, using the mathematics of symmetry to reduce the infinite ways that atoms can pack into materials into a finite set of possibilities. They then used machine learning to predict the ideal packing of atoms, given the elements and their relative composition in the material.

Their method accurately predicts the structure of materials that hold promise for piezoelectric and energy harvesting applications, with over five times the efficiency of current methods. Their method can also find thousands of new and stable materials that have never been made before, in a way that is computationally efficient.

“ ̽��ֱ��number of materials that are possible is four to five orders of magnitude larger than the total number of materials that we have made since antiquity,” said co-first author Dr Rhys Goodall, also from the Cavendish Laboratory. “Our approach provides an efficient computational approach that can ‘mine’ new stable materials that have never been made before. These hypothetical materials can then be computationally screened for their functional properties.”

̽��ֱ��researchers are now using their machine learning platform to find new functional materials such as dielectric materials. They are also integrating other aspects of experimental constraints into their materials discovery approach.

̽��ֱ��research was supported in part by the Royal Society and the Winton Programme for the Physics of Sustainability.

Reference:
Rhys A Goodall et al. ‘Rapid discovery of stable materials by coordinate-free coarse graining.’ Science Advances (2022). DOI: 10.1126/sciadv.abn4117

Researchers have designed a machine learning method that can predict the structure of new materials with five times the efficiency of the current standard, removing a key roadblock in developing advanced materials for applications such as energy storage and photovoltaics.

Our approach provides an efficient computational approach that can ‘mine’ new stable materials that have never been made before.

Rhys Goodall

MR.Cole_Photographer via Getty Images

Geometric abstract background with connected line and dots

Yes

AI techniques used to improve battery health and safety

sc604 — Mon, 06 Apr 2020 10:58:14 +0000

̽��ֱ��researchers, from Cambridge and Newcastle Universities, have designed a new way to monitor batteries by sending electrical pulses into them and measuring the response. ̽��ֱ��measurements are then processed by a machine learning algorithm to predict the battery’s health and useful lifespan. Their method is non-invasive and is a simple add-on to any existing battery system. ̽��ֱ��results are reported in the journal Nature Communications.

Predicting the state of health and the remaining useful lifespan of lithium-ion batteries is one of the big problems limiting widespread adoption of electric vehicles: it’s also a familiar annoyance to mobile phone users. Over time, battery performance degrades via a complex network of subtle chemical processes. Individually, each of these processes doesn’t have much of an effect on battery performance, but collectively they can severely shorten a battery’s performance and lifespan.

Current methods for predicting battery health are based on tracking the current and voltage during battery charging and discharging. This misses important features that indicate battery health. Tracking the many processes that are happening within the battery requires new ways of probing batteries in action, as well as new algorithms that can detect subtle signals as they are charged and discharged.

"Safety and reliability are the most important design criteria as we develop batteries that can pack a lot of energy in a small space," said Dr Alpha Lee from Cambridge’s Cavendish Laboratory, who co-led the research. "By improving the software that monitors charging and discharging, and using data-driven software to control the charging process, I believe we can power a big improvement in battery performance."

̽��ֱ��researchers designed a way to monitor batteries by sending electrical pulses into it and measuring its response. A machine learning model is then used to discover specific features in the electrical response that are the tell-tale sign of battery aging. ̽��ֱ��researchers performed over 20,000 experimental measurements to train the model, the largest dataset of its kind. Importantly, the model learns how to distinguish important signals from irrelevant noise. Their method is non-invasive and is a simple add-on to any existing battery systems.

̽��ֱ��researchers also showed that the machine learning model can be interpreted to give hints about the physical mechanism of degradation. ̽��ֱ��model can inform which electrical signals are most correlated with aging, which in turn allows them to design specific experiments to probe why and how batteries degrade.

"Machine learning complements and augments physical understanding," said co-first author Dr Yunwei Zhang, also from the Cavendish Laboratory. " ̽��ֱ��interpretable signals identified by our machine learning model are a starting point for future theoretical and experimental studies."

̽��ֱ��researchers are now using their machine learning platform to understand degradation in different battery chemistries. They are also developing optimal battery charging protocols, powering by machine learning, to enable fast charging and minimise degradation.

This work was carried out with funding from the Faraday Institution. Dr Lee is also a Research Fellow at St Catharine’s College.

Reference:
Yunwei Zhang et al. ‘Identifying degradation patterns of lithium ion batteries from impedance spectroscopy using machine learning.’ Nature Communications (2020). DOI: 10.1038/s41467-020-15235-7

Researchers have designed a machine learning method that can predict battery health with 10x higher accuracy than current industry standard, which could aid in the development of safer and more reliable batteries for electric vehicles and consumer electronics.

Nordwood Themes via Unsplash

Person holding white Android phone

Yes

AI learns the language of chemistry to predict how to make medicines

sc604 — Mon, 02 Sep 2019 23:00:01 +0000

̽��ֱ�� of Cambridge researchers have shown that an algorithm can predict the outcomes of complex chemical reactions with over 90% accuracy, outperforming trained chemists. ̽��ֱ��algorithm also shows chemists how to make target compounds, providing the chemical ‘map’ to the desired destination. ̽��ֱ��results are reported in two studies in the journals ACS Central Science and Chemical Communications.

A central challenge in drug discovery and materials science is finding ways to make complicated organic molecules by chemically joining together simpler building blocks. ̽��ֱ��problem is that those building blocks often react in unexpected ways.

“Making molecules is often described as an art realised with trial-and-error experimentation because our understanding of chemical reactivity is far from complete,” said Dr Alpha Lee from Cambridge’s Cavendish Laboratory, who led the studies. “Machine learning algorithms can have a better understanding of chemistry because they distil patterns of reactivity from millions of published chemical reactions, something that a chemist cannot do.”

̽��ֱ��algorithm developed by Lee and his group uses tools in pattern recognition to recognise how chemical groups in molecules react, by training the model on millions of reactions published in patents.

̽��ֱ��researchers looked at chemical reaction prediction as a machine translation problem. ̽��ֱ��reacting molecules are considered as one ‘language,’ while the product is considered as a different language. ̽��ֱ��model then uses the patterns in the text to learn how to ‘translate’ between the two languages.

Using this approach, the model achieves 90% accuracy in predicting the correct product of unseen chemical reactions, whereas the accuracy of trained human chemists is around 80%. ̽��ֱ��researchers say that the model is accurate enough to detect errors in the data and correctly predict a plethora of difficult reactions.

̽��ֱ��model also knows what it doesn’t know. It produces an uncertainty score, which eliminates incorrect predictions with 89% accuracy. As experiments are time-consuming, accurate prediction is crucial to avoid pursuing expensive experimental pathways that eventually end in failure.

In the second study, Lee and his group, collaborating with the biopharmaceutical company Pfizer, demonstrated the practical potential of the method in drug discovery.

̽��ֱ��researchers showed that when trained on published chemistry research, the model can make accurate predictions of reactions based on lab notebooks, showing that the model has learned the rules of chemistry and can apply it to drug discovery settings.

̽��ֱ��team also showed that the model can predict sequences of reactions that would lead to a desired product. They applied this methodology to diverse drug-like molecules, showing that the steps that it predicts are chemically reasonable. This technology can significantly reduce the time of preclinical drug discovery because it provides medicinal chemists with a blueprint of where to begin.

“Our platform is like a GPS for chemistry,” said Lee, who is also a Research Fellow at St Catharine’s College. “It informs chemists whether a reaction is a go or a no-go, and how to navigate reaction routes to make a new molecule.”

̽��ֱ��Cambridge researchers are currently using this reaction prediction technology to develop a complete platform that bridges the design-make-test cycle in drug discovery and materials discovery: predicting promising bioactive molecules, ways to make those complex organic molecules, and selecting the experiments that are the most informative. ̽��ֱ��researchers are now working on extracting chemical insights from the model, attempting to understand what it has learned that humans have not.

“We can potentially make a lot of progress in chemistry if we learn what kinds of patterns the model is looking at to make a prediction,” said Peter Bolgar, a PhD student in synthetic organic chemistry involved in both studies. “ ̽��ֱ��model and human chemists together would become extremely powerful in designing experiments, more than each would be without the other.”

̽��ֱ��research was supported by the Winton Programme for the Physics of Sustainability and the Herchel Smith Fund.

References:
Philippe Schwaller et al. ‘Molecular Transformer: A Model for Uncertainty-Calibrated Chemical Reaction Prediction.’ ACS Central Science (2019). DOI: 10.1021/acscentsci.9b00576

Alpha Lee et al. ‘Molecular Transformer unifies reaction prediction and retrosynthesis across pharma chemical space.’ Chemical Communications (2019). DOI: 10.1039/C9CC05122H

Researchers have designed a machine learning algorithm that predicts the outcome of chemical reactions with much higher accuracy than trained chemists and suggests ways to make complex molecules, removing a significant hurdle in drug discovery.

Our platform is like a GPS for chemistry

Alpha Lee

Денис Марчук from Pixabay

Background abstract line

Yes

Machine learning algorithm helps in the search for new drugs

sc604 — Mon, 11 Feb 2019 20:00:00 +0000

̽��ֱ��researchers, led by the ̽��ֱ�� of Cambridge, used their algorithm to identify four new molecules that activate a protein which is thought to be relevant for symptoms of Alzheimer’s disease and schizophrenia. ̽��ֱ��results are reported in the journal PNAS.

A key problem in drug discovery is predicting whether a molecule will activate a particular physiological process. It’s possible to build a statistical model by searching for chemical patterns shared among molecules known to activate that process, but the data to build these models is limited because experiments are costly and it is unclear which chemical patterns are statistically significant.

“Machine learning has made significant progress in areas such as computer vision where data is abundant,” said Dr Alpha Lee from Cambridge’s Cavendish Laboratory, and the study’s lead author. “ ̽��ֱ��next frontier is scientific applications such as drug discovery, where the amount of data is relatively limited but we do have physical insights about the problem, and the question becomes how to marry data with fundamental chemistry and physics.”

̽��ֱ��algorithm developed by Lee and his colleagues, in collaboration with biopharmaceutical company Pfizer, uses mathematics to separate pharmacologically relevant chemical patterns from irrelevant ones.

Importantly, the algorithm looks at both molecules known to be active and molecules known to be inactive and learns to recognise which parts of the molecules are important for drug action and which parts are not. A mathematical principle known as random matrix theory gives predictions about the statistical properties of a random and noisy dataset, which is then compared against the statistics of chemical features of active/inactive molecules to distil which chemical patterns are truly important for binding as opposed to arising simply by chance.

This methodology allows the researchers to fish out important chemical patterns not only from molecules that are active but also from molecules that are inactive – in other words, failed experiments can now be exploited with this technique.

̽��ֱ��researchers built a model starting with 222 active molecules and were able to computationally screen an additional six million molecules. From this, the researchers purchased and screened the 100 most relevant molecules. From these, they identified four new molecules that activate the CHRM1 receptor, a protein that may be relevant for Alzheimer’s disease and schizophrenia.

“ ̽��ֱ��ability to fish out four active molecules from six million is like finding a needle in a haystack,” said Lee. “A head-to-head comparison shows that our algorithm is twice as efficient as the industry standard.”

Making complex organic molecules is a significant challenge in chemistry, and potential drugs abound in the space of yet-unmakeable molecules. ̽��ֱ��Cambridge researchers are currently developing algorithms that predict ways to synthesise complex organic molecules, as well as extending the machine learning methodology to materials discovery.

̽��ֱ��research was supported by the Winton Programme for the Physics of Sustainability.

Reference:
Alpha A. Lee et al. ‘Ligand biological activity predicted by cleaning positive and negative chemical correlations.’ PNAS (2019). DOI: 10.1073/pnas.1810847116

Researchers have designed a machine learning algorithm for drug discovery which has been shown to be twice as efficient as the industry standard, which could accelerate the process of developing new treatments for disease.

̽��ֱ��ability to fish out four active molecules from six million is like finding a needle in a haystack

Alpha Lee

Yes