The heavily oversubscribed 3rd RSC-BMCS / RSC-CICAG Artificial Intelligence in Chemistry will taking place next week 28th-29th September 2020 Twitter hashtag - #AIChem20. There is an accompanying poster session and there is a chance to talk to the poster presenters in the breakout rooms at the end of each day (You will need the latest version of Zoom 5.3.0).
Most of the posters are now available for viewing on Twitter so you can always have a browse and ask questions on Twitter even if you won't be at the meeting #AIChem20poster,
Below is a table containing all posters.
|Poster Number||Name||Title||Twitter link|
|P01||Antreas Afantitis||Enalos cheminformatics tools: development of a de novo drug design module||View on twitter|
|P02||Nurlybek Amangeldiuly||Transfer learning with graph neural networks for protein_ligand binding kinetics prediction||View on twitter|
|P03||Andy Sode Anker||Characterising the atomic structure of mono_metallic nanoparticles from x_ray scattering data using conditional generative models||View on twitter|
|P04||Jenna Bilbrey||A look inside the black box: using graph_theoretical descriptors for the post_hoc interpretation of neural networks||View on twitter|
|P05||Nicolas Bosc||MAIP: a prediction platform for predicting blood_stage malaria inhibitors||View on twitter|
|P06||Xiaojing Cong||Receptor_ligand prediction by proteochemometric modeling: an application to G protein_coupled olfactory receptors||View on twitter|
|P07||Simon Durr||EVOLVE: a genetic algorithm to predict thermostability||View on twitter|
|P08||Umberto Esposito||Building a connected data pipeline to target drug development challenges|
|P09||Benedek Fabian||MolBERT: molecular representation learning with advanced language models and useful auxiliary tasks||View on twitter|
|P10||Miguel Garcia_Ortegon||Improving VAE molecular representations by tailoring them to predict docking poses and scores||View on twitter|
|P11||Wenhao Gao||Can we synthesize molecules proposed by generative models||View on twitter|
|P12||Helena Gaspar||Proteochemometric models using multiple sequence alignments and a SentencePiece_based masked language model: application to CYP and kinome selectivity modelling||View on twitter|
|P13||Ed Griffen||An explainable AI system for medicinal chemists||View on twitter|
|P14||Ed Griffen||"Chemists: AI is here, unite to get the benefits"||View on twitter|
|P15||Thomas Hadfield||Explicit incorporation of structural information into a fragment elaboration model via deep reinforcement learning||View on twitter|
|P16||Hans Hanley||"GENerateZ: designing anticancer drugs using transcriptomic data, genetic algorithms, and variational autoencoder"||View on twitter|
|P17||Fergus Imrie||Generating property_matched decoy molecules using deep learning||View on twitter|
|P18||Kjell Jorner||Uniform quantitative predictive modelling for route design||View on twitter|
|P19||Itai Levin||Computationally assisted synthesis planning for hybrid chemoenzymatic pathways||View on twitter|
|P20||Timur Madzhidov||Deep conditional variational autoencoder for reaction conditions prediction||View on twitter|
|P21||Gergely Makara||AI_assisted lead optimization with derivatization design||View on twitter|
|P22||Neann Mathai||Performance and scope of a similarity_based and a random forest_based machine learning approach for small_molecule target prediction||View on twitter|
|P23||Janosh Menke||Enhancing molecular fingerprints using neural networks||View on twitter|
|P24||Juan Carlos Mobarec||Evolutionary chemistry for the design of desired pharmacological profiles||View on twitter|
|P25||Rohit Modee||Neural network potentials for representing potential energy surface and their applicability for geometry optimization||View on twitter|
|P26||Joseph Morrone||Challenges and progress in combining docking programs with deep neural networks||View on twitter|
|P27||Eva Nittinger||Non_additivity in public and inhouse data and its influence on ML performance||View on twitter|
|P28||Ferruccio Palazzesi||Integrating multi task graph convolutional neural network with a deep generative model||View on twitter|
|P29||Yashaswi Pathak||Deep learning enabled inorganic material generator||View on twitter|
|P30||Quentin Perron||Integrating data_driven computer_aided synthetic planning with generative AI|
|P31||Daniel Probst||Classification of chemical reactions through NLP_inspired fingerprinting||View on twitter|
|P32||Mikolaj Sacha||Molecule edit graph attention network: modeling chemical reactions as sequences of graph edits||View on twitter|
|P34||Jenke Scheen||Data_driven estimation of optimally_designed perturbation networks in relative alchemical free energy calculations||View on twitter|
|P35||Philippe Schwaller||RXNMapper: unsupervised attention_guided atom_mapping||View on twitter|
|P36||Matthew Segall||Imputation versus prediction and applications in drug discovery||View on twitter|
|P37||Vishnu Sresht||Can generative models learn privileged substructures and identify new bioisosteres?||View on twitter|
|P38||Gergely Takács||Analysis of commercial and public compound databases by self_organizing maps||View on twitter|
|P39||Morgan Thomas||Towards integrating deep generative models with structure_based design||View on twitter|
|P40||Hao Tian||What is hidden behind allostery? An integrated framework to decipher key components in AuLOV dimerization||View on twitter|
|P41||Alain Vaucher||Learning how to do chemical reactions from data||View on twitter|
|P42||Alexander van Teijlingen||Beyond tripeptides _ two_step active machine learning for very large datasets||View on twitter|
|P43||James Wallace||eApps – enabling a predict first culture for computational medicinal chemistry||View on twitter|
|P45||Yuanqing Wang||Bayesian active drug discovery via deep graph kernel learning||View on twitter|
|P46||Robbie Warringham||DigitalGlassware: structuring and contextualising chemical outcomes for faster discovery||View on twitter|
|P48||Jerome Wicker||AIScape: a machine learning platform for activity and ADME predictions||View on twitter|
Some of the presenters have also recorded 2 min lightning presentation describing their work, these are available on the RSC CICAG YouTube channel.
Day 1 (odd numbered posters) https://youtu.be/ikbrrnjOvdc
Day 2 (even numbered posters) https://youtu.be/tf8cGodlWfU
The heavily oversubscribed 3rd RSC-BMCS / RSC-CICAG Artificial Intelligence in Chemistry will taking place next week 28th-29th September 2020 Twitter hashtag - #AIChem20.
This is now an online virtual event, this has required an immense amount of reorganisation behind the scenes and significant expenditure. This would not have been possible without the generous support of the event sponsors. AstraZeneca and MSD and the exhibitors CCDC, Concept Life Sciences, IKTOS, Liverpool ChiroChem, Mcule, o2h discovery.
Some of the exhibitors have kindly provided videos describing their work, why not have a browse.
If you are at the meeting they will also be able to talk to you directly at the breakout session at the end of each day.
The very popular chemical drawing package ChemDoodle has been updated.
ChemDoodle 2D v11.1 is a feature and stability update and is recommended for all users. The major new feature is advanced repeat unit support, both in graphical drawing and cheminformatics interpretation and expansion. Other notable features include improved reaction building interfaces and feedback, arc-length controls, and smaller EPS file output. This update also corrects an issue introduced by macOS 10.15.6, leading to corrupted and missing windows. So if you are experiencing these issues, make sure to install this update.
This could be a very interesting result.
Preprint on medrxiv A New Screening Method for COVID-19 based on Ocular Feature Recognition by Machine Learning Tools DOI.
… we found that the confirmed cases of COVID-19 present the consistent ocular pathological symbols; and we propose a new screening method of analyzing the eye-region images, captured by common CCD and CMOS cameras...
As ever, treat preprints with caution, but certainly well worth following.
VMD is a molecular visualization program for displaying, animating, and analyzing large biomolecular systems using 3-D graphics and built-in scripting. VMD supports computers running MacOS X, Unix, or Windows, is distributed free of charge, and includes source code. However, it still lacks a graphical user interface (GUI) for molecular manipulations when doing some modeling tasks.
A recent publication DOI describes work using the Tcl/Tk toolkit to develop a user-friendly GUI for VMD, named Molcontroller.
All the scripts and installation instructions are available https://github.com/ChenchenWu-hub/Molcontroller.