CCK-21: David Koes

TL;DR: CCK-21 will feature Prof. David Koes from the University of Pittsburgh on Friday, June 28th, 2024, at 4:00 pm in the Seminar Room, Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU. Free refreshments at 5 pm.

This Friday! We’re delighted to announce Prof. David Koes from the University of Pittsburgh will be speaking at our next Comp Chem Kitchen, CCK-21, on Friday, June 28th, 2024, at 4:00 pm in the Seminar Room, Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU. The title of his talk is ”Deep learning for structure-based drug discovery”.

Abstract

Prof. Koes will describe the training and development of convolutional neural networks for protein-ligand scoring and how these deep learning models are integrated into the GNINA molecular docking open source software. Successful prospective evaluations of GNINA will be discussed, including recent top performance in the Critical Assessment of Computational Hit-Finding Experiments (CACHE).  Additionally, he will describe his open source pharmacophore screening resource, Pharmit, which enables the screening of millions of compounds in seconds and discuss several generative approaches for hit discovery using deep generative models.

About Prof. David Koes

Prof. David R. Koes is an Associate Professor in the Department of Computational and Systems Biology at the University of Pittsburgh and an Associate Director of the Joint Carnegie Mellon-University of Pittsburgh Ph.D. Program in Computational Biology (CPCB). He is also affiliated with the Integrative Systems Biology, Intelligent Systems Program, and Computational Biomedicine & Biotechnology graduate programs at the University of Pennsylvania.

Prof. Koes develops novel computational algorithms and builds full-scale systems to support rapid and inexpensive drug discovery while simultaneously applying these methods to develop novel therapeutics. He seeks to unlock the power of computation and machine learning to solve challenging, real world problems and is a staunch advocate of open source software and open science.

CCK-19: Oxford + Cambridge + Berlin

Wednesday, 30 November 2022, 3:30-5 pm GMT

We are delighted to be returning with CCK-19 in the new seminar room of the Dorothy Crowfoot Hodgkin Building (a.k.a. the Phase II Biochemistry Department on Sherrington Road in the heart of the Science Area). We will be meeting jointly with our friends in Cambridge and Berlin via Zoom. We will have talks from Roger Sayle in Cambridge via Zoom, Matteo Ferla in Oxford, and Andrea Volkamer in Berlin, starting at 3:30 pm UK time. After the talks, ~5:15 pm, we will head over to The Royal Oak on Woodstock Road. More details below, and register to attend in person on Eventbrite or register for Zoom here: https://zoom.us/meeting/register/tJApf-uhrD0vH9S1U-9p6il24hB5RHtEh1ii.

Programme

Why Is It So Hard to Search Ultra-Large Chemical Libraries?

Roger Sayle, NextMove Software, Cambridge
https://www.nextmovesoftware.com/people.html

Fragmenstein: Stitching Compounds Together Like a Reanimated Corpse

Matteo Ferla, Oxford Protein Informatics Group, Department of Statistics, University of Oxford, Oxford
https://www.matteoferla.com/

Data-Driven Methods for Active Compound Design and Risk Assessment

Andrea Volkamer, Charité Berlin and Saarland University, Berlin
https://volkamerlab.org/

Locations

Please turn up 10 minutes before the start time to register where needed:

OxfordPhase II Biochemistry Seminar Room, New Biochemistry Building off South Parks Road

CambridgeCambridge Crystallographic Data Centre, Union Road

BerlinLecture Hall 1 (1st Floor), Institut für Pharmazie, Königin-Luisestrasse 2+4

After-event Locations

~5:15 pm GMT onwards:

OxfordThe Royal Oak, Woodstock Road

CambridgeThe Alma, Russell Court

BerlinLuise Dahlem, Königin-Luise-Straße

Please join us in person—just turn up at the Phase II Biochemistry Seminar Room in the Dorothy Crowfoot Hodgkin Building 10 minutes before the start time—or to join remotely please register here:

Register for Zoom: https://zoom.us/meeting/register/tJApf-uhrD0vH9S1U-9p6il24hB5RHtEh1ii

If you would like to present at a future meeting please just let us know (morris [] stats.ox.ac.uk). 🙂

In Person (Oxford, Cambridge or Berlin) & Virtual (Zoom) – Free & open to all
http://www.C-INF.net – https://CompChemKitchen.org/ – http://www.DigiDrug.NET

CCK-18

UPDATE: We are going virtual! Our next Comp Chem Kitchen, CCK-18, will be a webinar, on Friday, March 27, 2020, at 5-6 pm. We are delighted to announce that Prof. Andreas Bender from the University of Cambridge will be speaking, as well as Dr Vicky Hellon from F1000 Research. To join the webinar, you must sign up for a free ticket (limit 100). Sold out (as of March 23)! We are looking into streaming the event via YouTube. More info soon.

Prof. Andreas Bender

Dr. Andreas Bender (University of Cambridge): “AI in Drug Discovery – What is Realistic, What are Illusions?”

While ‘AI in Drug Discovery’ is currently a hot topic, it is important to be optimistic, yet realistic, about its opportunities in the area: Drugs work in biological systems, which are noisy by their very nature, and which we don’t understand properly. This presentation will outline approaches and applications of AI in the drug discovery field, with a particular focus on data currently available to us, and challenges with the translation of computational predictions into the physiological (in vivo) context.

Dr Andreas Bender is a Reader for Molecular Informatics with the Centre for Molecular Science Informatics at the Department of Chemistry of the University of Cambridge, leading a group of about 15 postdocs, PhD and graduate students and academic visitors in the area of life science data analysis. In parallel, he leads a team in Computational Drug Safety and ADME within the Clinical Pharmacology & Safety Sciences group at AstraZeneca in Cambridge (CPSS DSAI). In his work, Andreas is involved with the integration and analysis of chemical and biological data, aimed at understanding phenotypic compound action (such as cellular readouts, and also organism-level effects) on a mechanistic level, predicting molecular properties related to both compound effiacy and toxicity, as well as drug repurposing. He received his PhD from the University of Cambridge and worked in the Lead Discovery Informatics group at Novartis in Cambridge, MA, USA, as well as at Leiden University in the Netherlands before his current post. In 2013 he was awarded an ERC Starting Grant to model mixture effects of chemical structures in biological systems using mechanistic approaches, an area currently very little understood.

Dr Bender is a Reader for Molecular Informatics, at the Centre for Molecular Informatics in the Department of Chemistry, University of Cambridge; and Associate Director of Computational ADME/Tox (CPSS-DSAI), at AstraZeneca, Cambridge, UK,
 

Vicky Hellon (Associate Publisher, F1000Research): “F1000Research: Open Access publishing and Chemoinformatics

F1000 Research was established as a publishing platform in 2013 with the aim of pioneering open research through open data, open peer review and transparency throughout the publication process. The talk will discuss open publishing in general, how the publication model at F1000Research works and how this differs from more ‘traditional’ scientific publishing. There will also be discussion on the F1000Research Chemical Information Science Gateway and our experiences of publishing in the Chemoinformatics/software area.

Vicky Hellon is the Associate Publisher at F1000Research which she joined in 2017 after previous roles in Open Access publishing at Nature Research. Her role encompasses developing content across F1000Research, building collaborations and supporting the development of funder specific publishing platforms, particularly Wellcome Open Research and also HRB Open Research which she is currently project lead on. She is passionate about the progression of open research and particularly enjoys opportunities to actively engage with the research community.

We would like to thank the University of Oxford MPLS Network and Interdisciplinary Fund (NIF) for making CCK possible.

Richard Gowers from NextMove Software speaking at CCK-17

In addition to Wendy Warr and André Frade, we are pleased to announce Richard Gowers from NextMove Software will also be speaking at CCK-17:

Element Parsing Served Five Ways

Richard’s talk will explore different levels of optimising a seemingly simple task. Along the way, various algorithms used in chemical similarity, bioinformatics and text mining will be discussed.

Richard is a Senior Software Engineer at NextMove Software, which develops cheminformatics solutions deployed globally at 9 of the top 15 pharmaceutical companies. He is also the lead developer of MDAnalysis, a Python toolkit for… MD analysis, fancy that.

Congratulations to Dr Wendy Warr

2020 Herman Skolnik Award winner announced…

Dr. Wendy Warr will be the recipient of the 2020 Herman Skolnik Award presented by the ACS Division of Chemical Information, for her contributions to the fields of chemical information and a number of related fields that impinge on chemical information including chemical structure representation, substructure searching, retrosynthesis and reaction prediction.

The prize consists of a $3,000 honorarium and a plaque. Dr. Warr will also be invited to organize an award symposium at the Fall 2020 ACS National Meeting to be held in San Francisco.

For more than 40 years, Dr. Warr has had a global influence on chemical information and cheminformatics. She provides services to the chemical information, cheminformatics and computational chemistry communities worldwide, and has evolved into a key opinion leader and trend watcher. These trends have included: combinatorial chemistry, chemistry and the Internet, and intranet and ethernet in industry (these three in the early 1990s), outsourcing, changing pharma R&D strategies (these two in the early 2000s), and continue to today’s AI, cloud computing, and blockchain. 

She formed her current employment, Wendy Warr & Associates, in 1992. Since that time she has been successfully supplying business and competitive intelligence services to a broad spectrum of clients across the world. Her success stems from her extensive network, incredible energy, and deep curiosity, and her specialized market knowledge of chemical information, computational chemistry, drug discovery, cheminformatics, STM publishing, and scientific communication. Her clients have included at least 15 major pharmaceutical and chemical companies, venture capitalists, financial analysts, all of the well-known chemistry publishers, software companies, and many cheminformatics and analytical chemistry companies, as well as many smaller commercial and not-for-profit or academic organizations. Scientific database producers have benefited from her expert counsel and services in recent years.

Dr. Warr obtained her doctoral degree (D. Phil.) from the University of Oxford in 1971, and subsequently joined ICI Pharmaceuticals where she held multiple positions culminating in leading the department of Information Services. In 1992, Dr. Warr established her own company, Wendy Warr & Associates, that provides business and competitive intelligence services to a broad spectrum of clients in the United States, Europe, Australia, the Middle East, and Asia.

She has played key roles in several professional organizations including the American Chemical Society, Royal Society of Chemistry, German Chemical Society, Society of Chemical Industry, Chemical Structure Association Trust  and the Institute of Information Scientists, and in many cases has been instrumental in shaping their activities. 

She has published over 80 articles in academic journals and over 100 commercial reports, along with numerous invited lectures at venues such as NIST, Washington DC and University of Strasbourg, France. She has been an Associate Editor for the Journal of Chemical Information and Modeling (as well as its predecessor, the Journal of Chemical Information and Computer Sciences).  She has received numerous awards and honors including the Ernie Hyde Award of the Chemical Structure Association (1984), and is a Fellow of the Royal Society of Chemistry and a Fellow of the Chartered Institute of Library and Information Professionals.

Raajarshi Guha
Chair, CINF Awards Committee

Congratulations, Wendy! We are looking forward to hosting you in Oxford for CCK-17.

CCK-17

TL;DR: CCK-17, will be on Thursday, November 28th, 2019, at 5:00 pm in the Seminar Room, Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU. Free CCK-17 tickets are available.

Dear Friends and Colleagues,

Dr Wendy Warr (Wendy Warr and Associates, UK): “My Way”

“Comp Chem Kitchen”, CCK-17, will be at 5-6 pm on Thursday, November 28th, 2019, in the Seminar Room, Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU. Our main talk will be a little different from our usual talks. We are delighted to announce Oxford Chemistry alumna, and just-announced winner of the 2020 Herman Skolnick Award, Dr Wendy Warr will be speaking:

Dr Wendy Warr
  • I would like to run through my very varied career experiences but then relate them to what I have done much more recently, and, ultimately, what current students might do in future. For example, the introduction of high throughput screening (HTS) in about 1989 led me to travel around the United States looking at robotically controlled compound stores. I have always had a particular interest in chemical reaction handling and that relates to the current revival in interest in retrosynthesis software, and to AI for reaction prediction. In about 1993-1996, I became a combinatorial chemistry guru. That relates to robotic synthesis also. My very early interest in chemical structure representation (1968!), and later in graphics-based systems for structure searching, relates to the current interest in huge databases of virtual molecules (e.g., Enamine REAL), big data, the need for ultrafast structure searching systems, and de novo drug design.

CAREER HISTORY

  • January 1992 – present, President, Wendy Warr & Associates
    • Wendy Warr & Associates supplies business and competitive intelligence services to a broad spectrum of clients in the United States, Europe, Australia, the Middle East, and Asia. Our success stems from our extensive network and our specialized knowledge of publishing, cheminformatics, and computational chemistry. Pharmaceutical companies, venture capitalists, publishers, software companies, and scientific database producers have benefited from our expert counsel and services in recent years.
  • 1972 – 1991, ICI Pharmaceuticals
    • 1986 – 1992, Manager, Information Services (managed department of 50+ people)
    • 1984 – 1986, Manager, Research Information
    • 1979 – 1984, Senior Systems Analyst/Project Manager for Chemical Information
    • 1976 – 1979, Senior Information Scientist and Systems Analyst
    • 1972 – 1976, Information Scientist
  • 1970 – 1972, Robinson Brothers Ltd., Research Chemist
  • 1968 – 1970, University of Oxford, Part-time Research Assistant, Experimental Information Unit

Lightning Talks

  • Richard Gowers (Senior Software Engineer, NextMove Software):Element Parsing Served Five Ways
    • This talk will explore different levels of optimising a seemingly simple task. Along the way various algorithms used in chemical similarity, bioinformatics and text mining will be discussed.
      • Richard is a Senior Software Engineer at NextMove Software, which develops cheminformatics solutions deployed globally at 9 of the top 15 pharmaceutical companies. He is also the lead developer of MDAnalysis, a Python toolkit for… MD analysis, fancy that.
  • Andre Frade (DPhil student, Department of Chemistry):From molecular diagrams to material properties: development of new tools and frameworks”.

Refreshments will be provided, including beer.

We would like to thank the University of Oxford MPLS Network and Interdisciplinary Fund (NIF) for making CCK possible.

Continue reading

CCK-16

TL;DR: CCK-16, the next in our series of meetings for computational chemists, cheminformaticians, and molecular modelers, will be on Thursday, October 31st, 2019, at 5:00 pm in the Seminar Room, Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU. Free CCK-16 tickets are available.

Dear Friends and Colleagues,

Please join us for our next “Comp Chem Kitchen”, CCK-16, at 5-6 pm on Thursday, October 31st, 2019, in the Seminar Room, Department of Biochemistry, University of Oxford, South Parks Road, Oxford, OX1 3QU. We are very pleased to announce Dr Paulo Tosco from Cresset, UK, will be speaking:

Dr Paulo Tosco

Dr Paulo Tosco (Cresset, UK): “Design in 2D, Model in 3D: Live 3D Pose Generation from 2D Sketches”

How nice would it be to be able to draw a molecule in your favourite 2D sketcher, and see in real time how its 3D electrostatic potential looks like, what interactions it makes in the active site, how its shape and electrostatics compare to a known active? This is why we decided to build from ground up a new molecular design platform around this engaging idea. As we started fleshing out the details of the Grow3D algorithm, a number of obvious and more subtle gotchas emerged.

In this talk I will illustrate how we addressed these challenges, and I will also give an overview of the integration of Grow3D in the context of our web-based molecular design platform, showing its effectiveness in bridging the gap between 2D and 3D methods.

Dr Paolo Tosco joined Cresset in 2014 after being an Assistant Professor at the University of Turin, Italy. A medicinal chemist by background, after completing his PhD in Drug Chemistry his main focus became computational chemistry, with a particular interest in development and implementation of novel methods and algorithms. He has been a contributor to many open source projects, most notably the RDKit. His current role at Cresset involves scientific development of the company codebase and technical customer support.

Lightning Talks

  • Mihaela Smilova (SGC):  “Fragment Hotspot Maps to Drive the Elaboration of Fragment Screening Hits“; and
  • Dr Tomasz Piskorz (Department of Chemistry):A Short Intro to MDAnalysis (A Python Tool to Analyze the Results of MD simulations)”.

Refreshments will be provided, including beer.

We would like to thank the University of Oxford MPLS Network and Interdisciplinary Fund (NIF) for making CCK possible.

Continue reading

CCK-15

TL;DR: CCK-15, the next in our series of meetings for computational chemists, cheminformaticians, and molecular modelers, will be on Thursday, September 5th, 2019, at 5:30 pm in the Large Lecture Theatre LG.01, Department of Statistics, University of Oxford, 24-29 St Giles’, Oxford OX1 3LB. Free CCK-15 tickets are available.

Dear Friends and Colleagues,

Please join us for our next “Comp Chem Kitchen”, CCK-15, at 5:30-6:30 pm on Thursday, September 5th, 2019, in the Large Lecture Theatre LG.01 in the Department of Statistics, University of Oxford, 24-29 St Giles’, Oxford. We are  very pleased to announce Prof. Adrian E. Roitberg from the Department of Chemistry, University of Florida, USA, will be speaking:

  • Prof. Adrian E. Roitberg (Department of Chemistry, University of Florida. Gainesville, FL. 32611-7200, USA; [email protected]):Is Quantum Chemistry Amenable for Machine Learning? Are the Computers Coming for Our Jobs?
    • For decades, computational chemists have been searching for new ways to develop faster and more accurate empirical potentials for application in a myriad of problems. These applications include the calculation of thermochemical properties and theoretical IR spectra, stability comparisons, structural optimization, and molecular dynamics or Monte Carlo simulation. The reason empirical potentials have been so attractive to the community abroad is the promise of accuracy and speed, especially compared to more traditional quantum mechanical or classical methods for computing molecular energies and forces. However, empirical potentials developed using traditional methods have remained at arm’s length for various reasons. Within the last decade, Machine Learning (ML) methods have proved applicable to the development of empirical potentials, specifically methods from deep learning. The ANAKIN-ME (ANI) method, which focuses on the use of large and diverse data sets in training new potentials, has consistently proven to be universally applicable to systems containing the atomic species in the training set. Focusing on parametrization for organic molecules, we have developed a universal ANI potential which is highly accurate compared to reference QM calculations at speeds 107 faster. The ANI potential is shown to accurately represent the underlying physics of molecules through various test cases including: thermochemistry, structural optimization, and molecular dynamics simulations. The results presented here provide evidence of the universal applicability of ANI deep learned potentials to various computational chemistry problems involving organic molecules.

Prof. Roitberg was born and raised in Argentina, and he received his BS degree in Chemistry from the University of Buenos Aires. He then moved to the US to pursue his PhD in chemistry, at the University of Illinois at Chicago, followed by a postdoctoral stay at Northwestern University. He joined a National laboratory (NIST) as staff and moved to the University of Florida as Assistant Professor in 2001. He is now a Full Professor there at UF, in both the Chemistry and Physics Departments. Prof. Roitberg has graduated 20 PhD students so far. He has received several awards, including being named a Fellow of both the American Physical Society and the American Chemical Society. He was also named an Ulam Scholar at Los Alamos National Laboratory and received the Raices Prize (roots) from the Ministry of Science in his native Argentina. Prof. Roitberg has published over 160 scientific articles, and his h-index is 55, with over 24000 citations of his papers.

Lightning Talks

  • Harry Morgan (Physical and Theoretical Chemistry Laboratory. South Parks Road, Oxford, OX1 3QZ, UK): Pressure-induced transitions in vanadium oxyhydrides“.
  • Dr Anthony Nash, PhD, MRSC (Senior Research Scientist, Nuffield Department of Clinical Neurosciences, University of Oxford): Human Centric Drug Discovery Using a Real World Dataset and iPSC-based Assays”.

Refreshments will be provided, including beer.

We would like to thank the University of Oxford MPLS Network and Interdisciplinary Fund (NIF) for making CCK possible.

Want to speak? Ideas for speakers?

If you have ideas for speakers, or would like to give a talk, get in touch. We also invite lightning talks of 5 minutes (or fewer) from attendees, so if you have some cool code you’ve been working on and would like to demo it, bring your laptop, smartphone, tablet, (wearable?) and tell us all about it.

The talks usually resemble one of the following styles:

  • an overview of computational chemistry in your research;
  • a (live!) demonstration of some software that you are developing or using; or
  • a summary of a computational chemistry paper, method, programming language, or tool that you’ve seen recently.

Please pass this announcement on to friends, colleagues, and students who may be interested too!

About CCK

Comp Chem Kitchen is a regular forum and seminar series to hear about and discuss computational methods for tackling problems in chemistry, biochemistry and drug discovery. It focuses principally on cheminformatics, computational chemistry, and molecular modelling, and overlaps with neighboring areas such as materials properties and bioinformatics.

We’re keen to encourage people involved in coding and methods development (i.e. hackers, in the original untarnished sense of the word) to join us. Our hope is that we will share best practices, even code snippets and software tools, and avoid re-inventing wheels.

In addition to local researchers, we invite speakers from industry and non-profits from time to time, and occasionally organize software demos and tutorials.

If you’re interested in giving a talk, here are some possible topics:

  • Software development (e.g.: Python, C, C++, CUDA, shell, Matlab);
  • Optimizing force field parameters & EVB models;
  • Cheminformatics (e.g.: RDKit);
  • X-ray and NMR crystallography, including small molecule and macromolecular;
  • Protein & RNA modeling, including Molecular Dynamics;
  • Virtual screening and Docking;
  • Machine Learning;
  • Quantum Methods, including DFT.

Bring your laptops, by the way, if you have something you’d like to show!

The main CCK web site is: http://compchemkitchen.org/
Follow us on Twitter: @CompChemKitchen
See you soon! We’re looking forward to seeing and hearing about the diverse range of computational molecular science that you’re cooking up…

—Garrett, Richard, Phil, and Fernanda

[email protected]
[email protected]
[email protected]
[email protected]