Event Date/s: From Friday 02 October 2020 - 09:30
To Friday 09 October 2020 - 18:00
This is a week long series of workshops (mixtures of lectures and hands-on activities) and seminars aimed at newcomers to biomolecular simulations and those who wish to improve their skills. It is designed in a modular way so registered participants can pick and choose the sessions that suit their individual interests. We recommend installing the relevant software before each session (if you are unsure about this attending the Introduction to Software session would be highly encouraged). This webpage will be updated with any new information.
This is an online event. Approximately one week before the final schedule and links will be sent to registered participants. All times will be in the British time zone.
Registration is now open. Please register for this workshop on the event booking site.
Registration Deadline: 25 September 2020
Space Available - 1000
Fee - none
Monday 5 October 16:00 - 17:00 Syma Khalid, University of Southampton : Towards a virtual bacterial cell envelope: adding the biological complexity to biomolecular MD simulations.
Abstract: Gram-negative behaviour are protected cell envelopes composed of two membranes separated by an aqueous region known as the periplasm. Here I will describe our latest work in which we use complex models of bacterial cell envelopes to perform atomistic molecular dynamics simulations. In particular I will share very recent results in which new hypotheses are generated from unexpected, but in hindsight, intuitive behaviour when realistic crowding is incorporated into the models.
Thursday 8 October 16:30 - 17:30 Prem Chapagain, Florida International University : Using molecular dynamics simulations to design novel inhibitors of the topoisomerase class of DNA processing enzymes
Abstract: Newtonian mechanics of particle motion can provide unprecedented details and complex relationships among molecular structure and function. With recent advances in the computing speed as well as vast improvements in the force field, many-particle molecular dynamics simulation has resulted in accurate predictions of biomolecular structure and dynamics. This development has marked a significant turning point in the physics-based investigations of the molecules of life. Today, MD has found numerous applications, ranging from structure refinement of low-resolution crystal data to computer-aided drug discovery efforts in treating a myriad of diseases. In this talk, I will discuss the use of molecular simulation in several molecular systems with special focus on the simulations and docking to identify novel Topoisomerase I inhibitors, explaining how in silico screening targeting an active site pocket of MtbTopI can allow identification of top hits, which could be further explored with inhibition assays of MtbTopI activity.
Friday 9 October 16:30 - 17:30 Viv Kendon, Durham University : Quantum Computing for biomolecular simulations: Current status and future prospects [with CCP-QC]
Abstract: You have probably noticed that new computers are not getting faster, but HPC facilities are becoming more diverse. Dedicated hardware for neural networks and machine learning; GPUs; FPGAs; ASICs; each requiring new skills to programme and integrate with conventional computers. And now, quantum computing is promising to be the latest and greatest innovation to speed up your computations. Computational scientist know that it is never that easy, a lot of hard work goes into simulation advances by any route. There are good theoretical reasons to support the idea that quantum computers can eventually make a bigger difference to many simulations than other types of new hardware. To facilitate the process of integrating quantum computing into scientific computing, CCP-QC was proposed and funded in the recent call for CCPs. I will give an overview of CCP-QC's plans to work with existing CCPs, HECs, and the quantum computing community. This will include an introduction to current state of the art in quantum computing, and some insights into what we expect will work (and what won't) and why.
Our training programme has been colour coded to help you decide which sessions are best for you. Green is easy, blue is intermediate and red is advanced. Please find more details here. The final schedule will be sent to all registered participants.
9:30 - 12:30 Introduction to Software Installation for Biomolecular Simulations [James Gebbie-Rayet]
This talk will provide background information about what software runs where and how to set up your computer to run the programs you need. It will be followed by a hands-on session where students can try installing software, ask questions, and get help troubleshooting.
Monday 5 October 2020
9:30 – 10:30 The Basics of Biomolecular Simulations [Sarah Harris]
This lecture will cover the fundamental concepts of biomolecular simulations, with a focus on atomistic molecular dynamics (MD).
10:45 – 12:30 Seeing Biomolecules with VMD [Sarah Harris]
I will demonstrate how to use the programme Visual Molecular Dynamics (VMD) to visualise biomolecular structures from the protein data bank and from molecular dynamics simulations. You need to have installed the VMD software.
Training material can be found here.
13:30 – 15:30 Setting up an AMBER Simulation [Sarah Harris]
I will demonstrate how to set up a simulation of a protein and a protein-ligand complex using the AMBER suite of programs. You need to have installed AmberTools and VMD would be useful.
Training material can be found here.
Tuesday 6 October 2020
9:30 – 12:30 Trusting Your Simulations [Charlie Laughton]
Whether you are a user of Amber, Charmm, Gromacs, NAMD or any other MD package, there are a good number of on-line tutorials that will take you through the mechanics of setting up and running an MD simulation. However, in general there is less discussion in these about how to ensure you end up with a good simulation. In this workshop we will explore some of the issues in simulation preparation and analysis that can trip up the unwary, and how to avoid them. The hands-on parts of this tutorial are run from the training page of the CCPBioSim website so you only need a web browser installed. Please register on the CCPBioSim training page before this workshop.
13:30 – 15:30 Structural Bioinformatics of Variants: Principles and Tools to Study Genetic Variants in the Context of Protein Structure [Franca Fraternali]
The classification of missense variants in terms of their consequences to protein stability and pathogenicity is a very complex task. The presence of variants do not necessarily indicate pathogenicity: genetic variants are present amongst the healthy population, but some of them are causative of human diseases. Typically, sequence conservation and variant incidence are used as features in variant impact predictors. These often prove to be not conclusive to discriminate damaging from non-damaging loci. It is becoming apparent that protein structure and dynamics derived features, when available, could contribute to variants’ impact prediction. The biological community has begun to map out other levels of biological complexity using high-throughput approaches, generating measurements of transcript and protein abundance, and interactions between these entities. In this workshop we will introduce principles and methods in studying the impact of genetic variants on protein structures. We will discuss the mapping of variations to protein structural data, protein-protein interaction sites, and in silico approaches to exploit these data types to annotate variant impact. Participants will have hands-on practices to navigate web-based resources, tools and methods to acquire and map genetic variant data, and understand the impact of variants pose on protein structure and dynamics. This workshop will use ZoomVar and you need to have installed PyMOL and Jalview.
16:00 – 17:00 Validating Biomolecular Structures with ISOLDE [Tristan Croll]
While the worldwide Protein Data Bank (wwPDB) is an amazing resource encapsulating the results of four decades of experimental structural biology, many of the structures therein come with a hefty dose of "caveat emptor". While methods have of course improved over time, residual errors of varying levels of severity remain the norm in deposited models. While some of these mistakes are of the trivial variety that will correct themselves with a little energy minimisation, many remain beyond the scope of automated correction. In this presentation I will demonstrate how ISOLDE (an interactive molecular dynamics based model-building plugin to UCSF ChimeraX - see https://isolde.cimr.cam.ac.uk) can be used to interactively inspect, correct (or judge as intractable) a model against its original experimental map(s). For most standard cases, minimal knowledge of the theory underlying x-ray crystallography or cryoEM is required.
Wednesday 7 October 2020
9:30 – 12:30 and 13:30 – 15:30 Docking with HADDOCK [Alexandre Bonvin]
This session will start with talks covering protein-ligand modelling, antibody-antigen modelling, HADDOCKing with MS data, HADDOCKing with cryo-EM data, and Visualization of the interaction space defined by restraints. This will be followed by a hands-on tutorial using HADDOCK. You need to have installed PyMOL.
16:00 – 17:00 How to Interpret Your Docking Scores: A Medicinal Chemistry Perspective [Geoff Wells]
This presentation will discuss how to interpret the output from in silico docking experiments, particularly relating to their chemistry, physicochemical properties and known biological activities. We will explore the use of approved drugs for potential repurposing and also fragment and small molecule libraries for the de novo design of new ligands. We will consider properties we should look for and avoid in our 'virtual hits' and suggest some next steps for refining and filtering hit compounds, whilst acknowledging that medicinal chemists can have very different views on what is the ‘right’ answer. We will briefly discuss the use of filters based on binding energy, ligand efficiency, Lipinski's rules, toxicophores, PAINS, reactivity, and synthetic or commercial accessibility. The goal is to propose a group of tools and resources and some selection criteria that can be used during the evaluation, processing and use of ligand docking results. Suggested open source docking tools and resources include: FTMAP (webserver), Autodock Vina (various versions), secondary scoring functions (e.g. Cyscore), visualisation tools such as UCSF Chimera, and the ZINC compound database and pharmacophore tools (ZINCPharmer).
Thursday 8 October 2020
9:30 - 12:30 BioSimSpace in Jupyter [Christopher Woods / Lester Hedges]
This workshop aims at giving a broad introduction of the functionalities of BioSimSpace, the flagship software being produced in partnership with CCP-BioSim/HEC-BioSim. BioSimSpace provides a Python API to commonly used simulation packages such as Gromacs, Amber and NAMD that allows the setup and running of complex molecular simulations as well as their analysis. The workshop will cover parameterisation of small molecules and proteins, solvation, minimisation and equilibration of the molecular systems. It will look at how to visualise molecular systems, as well as plot observables of running simulations in the Jupyter notebook environment to track their progress.
13:30 – 14:30 Materials Modelling Force Fields: Successes and Limitations [Alin Elena] [with CCP5]
This short introductory lecture about force fields will provide an overview on the most popular existing force fields available, their advantages and shortcomings.
14:40 – 16:00 When Cryo-EM Meets Biomolecular Simulations [Tom Burnley] [with CCP-EM]
This lecture will introduce cryo-electron microscopy discussing background, different modalities (tomography vs single particle), data processing, future direction and potential areas for interfacing with molecular dynamics.
Friday 9 October 2020
9:30 – 12:00 Large Scale Modelling with DL_MESO [Michael Seaton] [with CCP5]
Modelling large-scale molecular systems is dependent on getting both correct thermodynamic and hydrodynamic behaviours, which can be achieved using mesoscale modelling techniques that operate between atomistic and continuum-based length and time scales. DL_MESO is a general-purpose mesoscale modelling simulation suite, consisting of highly scalable codes for two mesoscopic methods: Dissipative Particle Dynamics (DPD) and the Lattice Boltzmann Equation (LBE). This workshop will introduce these two mesoscale modelling techniques and DL_MESO, with practical examples and details of how they can be used to model biological/biomolecular systems. Software required: DL_MESO; VMD or OVITO; Paraview; a data plotting program. Some optional python3 scripts will be provided.
13:00 – 16:00 Modelling Enzymes with QM/MM [Marc van der Kamp]
The training workshop will introduce non-specialists to the use of combined quantum mechanics/molecular mechanics (QM/MM) methods for modelling enzyme-catalysed reaction mechanisms. Concepts and techniques of QM/MM reaction modelling will be explained through hands-on exercises. During the tutorial, each participant will generate and analyse a free energy profile and a potential energy profile for the reaction catalysed by chorismate mutase. The hands-on parts of this tutorial are run from the training page of the CCPBioSim website so you only need a web browser installed. Please register on the CCPBioSim training page before the workshop.
For further information or queries, contact: