Artificial intelligence (AI) – Page 3
-
FeatureWhat does AI mean for chemistry?
Phil Ball looks at whether letting machines do our thinking for us will change our understanding of chemistry itself
-
OpinionThe psychology of our future with AI
It’s time to accept that digitalisation is changing laboratory work, and embrace the opportunity
-
ResearchAI model accurately classifies reaction mechanisms
Machine learning ‘surpasses chemist experts’ in identifying chemical processes
-
OpinionHow long until papers list an AI as an author?
Artificial intelligence is superior to humans at numerous tasks, but it is still vulnerable to human biases
-
ResearchAlgorithm produces one of the best solutions to molecules’ Schrödinger equations yet
But aggressive ‘pre-training’ actually damages its predictive powers
-
OpinionExploring AlphaFold’s knowledge of energy landscapes
The algorithm needs a little help to find the global energy minimum
-
ResearchAI beats human experts when it comes to peptide design
Algorithm throws up some counter-intuitive structures in contest
-
ResearchAI discovers the best general conditions yet for cross couplings, doubling yields
Algorithm works with robotic experimenter on tricky Suzuki–Miyaura reactions
-
NewsAI picks out fake science
Textual analysis tool flags manuscripts that may have come from paper mills
-
OpinionWhy AlphaFold won’t revolutionise drug discovery
Protein structure prediction is a hard problem, but even harder ones remain
-
NewsAlphaFold has predicted the structures of almost every known protein
Google offshoot DeepMind has released more than 200 million predicted 3D structures, covering nearly the entire protein universe
-
NewsYield-predicting AI needs chemists to stop ignoring failed experiments
For machine-learning algorithms to make accurate yield predictions, chemists need to start reporting more low-yielding reactions and spell out implied experimental details
-
ResearchAI-engineered enzyme eats entire plastic containers
Enzyme with only five amino acid alterations depolymerises 51 different PET products faster and at lower temperature than other proteins
-
ResearchAI tool finds green ways to turn chemical waste into drugs
Circular approach for industry by-products considers environmental, economic and geographic aspects to rank predicted routes for real-world applications
-
NewsDrug discovery AI that developed new nerve agents raises difficult questions
Researchers shocked by how easily algorithm discovered novel poisons even deadlier than VX
-
ResearchMolecular counterfactuals method helps researchers explain AI predictions
Understanding machine learning predictions by exploring the road not travelled
-
OpinionThe chemist’s gambit
If artificial intelligence can revolutionise chess, what might it do to chemistry?
-
OpinionThe law of conservation of data
AI and machine learning are useful and powerful, but they need high quality data inputs that aren’t available yet for drug discovery
-
ResearchAlgorithm out of Google’s DeepMind finesses DFT calculations
Machine learning creates algorithm that avoids large errors in solutions to certain problems
-
ResearchUnsupervised machine-learning tool could accelerate catalyst discovery
The approach was able to identify phosphine ligands that may form dinuclear palladium(I) complexes using only five experimental data points