About BAGIM

BAGIM is an active community of Boston area scientists bringing together people from diverse fields of modeling and informatics to impact life and health sciences. BAGIM strives to create a forum for great scientific discussions covering a wide range of topics including data management, visualization, computational chemistry, drug discovery, protein structure, molecular modeling, structure-based drug design, data mining, software tools, and the sharing of goals and experiences. Our community is made up of participants from academia, government, and industry whose goal is to engage in the discussion of science involving a synthesis of theory and technology. Discussions sponsored by BAGIM are targeted to the needs and interests of informatics scientists, computational chemists, medicinal chemists, and statisticians. BAGIM also provides opportunities for networking within these disciplines as well as an arena for the dissemination of information of specific interest to the membership.

Friday, December 15, 2017

BAGIM Event: Dr. Douglas Selinger at Merck in Boston Thursday Jan 11 2018 starting at 6pm

Dr. Douglas Selinger, CEO of Plex Research, will present "A search for cures: Chemical biology data analysis using search algorithms" on January 11th, 2018. The presentation will start at 6pm at the Merck Research Laboratories at 33 Avenue Louis Pasteur in Boston. Please arrive a few minutes early. Networking and refreshments will be after the talk in the same location.

Please RSVP for this meeting

Abstract

We can find web sites. We can find restaurant reviews. Yet our scientific data and algorithms are buried in databases and spread across incompatible sources. Why not build a search engine for science?

Search engine algorithms, originally designed to make sense of interconnected web pages, can be repurposed to analyze linked data of all kinds. At Plex we’ve developed a search engine which incorporates chemical similarity (e.g. 2D/3D fingerprints), biological fingerprints (e.g. transcriptional profiling), and machine learning approaches. We’ve pointed this engine at a large corpus of publicly available data, including: compounds, targets, pathways, biomarkers, ADME, toxicity, diseases, patents, publications, and more. The results are powerful, yet simple.

Come join us to learn more as we explore the uses of search engine algorithms in drug discovery.


BAGIM is sponsored by Accenture, Dassault Systemes, DNASTAR, OpenEye Scientific, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Dotmatics, Cresset, Cyrus Biotechnology, Optibrium, Team Arrayo, and Scilligence

Wednesday, October 18, 2017

BAGIM Event: Dr Gabriel Musso at the Broad Institute in Cambridge Thursday November 16, starting at 6pm

Dr. Gabriel Musso, the CSO of BioSymetrics in Toronto, will present "Using Machine Learning to Screen Small Molecules" on November 16th. The presentation will start at 6pm at the Broad Institute. Please use the 415 Main Street entrance and arrive a few minutes early. Networking and refreshments will follow at around 7pm at Meadhall, 4 Cambridge Center, Cambridge. Please RSVP for this event.

Abstract

Drug discovery in the post-genomic era has largely focused on modulation of disease-relevant drug targets in hypothesis-driven screens. However, despite tremendous achievements, there is substantial pressure within the industry to address the high attrition rate associated with drug development. Failures in the clinical development of lead compounds may arise from lack of efficacy, unexpected toxicity, or anomalies in absorption, distribution, metabolism, or excretion. One potential solution to this problem is to incorporate an initial phenotype-based screen able to assess multiple specificity and toxicity phenotypes simultaneously. In this talk I’ll begin by describing previous work in which we performed phenotype-based screening of a library of bioactive compounds, using the results to generate machine learning models that prioritized as-yet-untested compounds for specific phenotypic effects. These models were used to generate nearly 700 million predicted structure-activity relationships, and highlight a mechanism-agnostic means for combined in vivo and in silico screening of massive chemical libraries. Our current goal is to design an informatics suite that would produce more advanced small molecule activity predictions by incorporating experimental data from multiple sources. We feel this approach has the ability to disrupt traditional lead compound discovery frameworks.

BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, OpenEye Scientific, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Dotmatics, Cresset, Cyrus Biotechnology, Optibrium, Team Arrayo, and Scilligence

Thursday, September 14, 2017

BAGIM Event: Dr. Heidi Koldsø at Moderna in Cambridge, Thursday Oct 19th

Dr. Heidi Koldsø from D. E. Shaw Research will present "Permeation, gating, and modulation of the TRPA1 channel in long-timescale molecular dynamics simulations"on Thursday, October 19, 2017.  The talk will be held at Moderna Therapeutics at 200 Technology Square in Cambridge, starting at 6pm.  Please plan to arrive ten to fifteen minutes before the start.  Networking and refreshments will follow the talk at Meadhall, 4 Cambridge Center. 

TRPA1, a member of the transient receptor potential (TRP) ion channel superfamily, is a nonspecific cation channel activated by natural chemicals found in, for instance, mustard, oregano, and cinnamon.  TRPA1 has recently appeared as a promising drug target in pain and inflammatory diseases.  Structure-based design efforts for this protein are problematic, however, due to the lack of molecular details of relevant channel functional states and of channel activation. 


To address this problem, we have studied TRPA1 channel functional states and channel activation using long-timescale, all-atom molecular dynamics simulations enabled by special-purpose hardware.  Specifically, we simulated inward and outward ionic currents, which helped us identify possible key structural determinants of channel gating.  We also studied modulation of the open and conducting channel by performing simulations of free binding of the TRPA1 inhibitor A-967079 to TRPA1.  Intriguingly, A-967079 was observed to bind in a location previously hypothesized to be a binding site.  Our observations suggest that this site may indeed be a target for development of new allosteric modulators, an avenue we intend to explore in the future.

Heidi Koldsø is a member of the technical team at D. E. Shaw Research, a company that develops specialized hardware and software for basic and applied computational biochemical research. She holds a Ph.D. in chemistry from Aarhus University in Denmark, as well as a M.Sc. and B.Sc. in Medicinal Chemistry, also from Aarhus University. After her Ph.D., she was a postdoctoral researcher at the Department of Biochemistry, University of Oxford. At D. E. Shaw Research, she applies molecular dynamics simulations to study biological systems, with a focus on ion channels.

BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, OpenEye Scientific, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Dotmatics, Cresset, Cyrus Biotechnology, Optibrium, Team Arrayo, and Scilligence

Monday, September 11, 2017

BAGIM Event: Panel on Cloud Computing for Molecular Modeling at NIBR in Cambridge, Monday September 11, 2017

A panel of industry experts will present a discussion on Cloud Computing for Molecular Modeling in the auditorium at NIBR at 250 Mass Avenue in Cambridge, starting at 6pm, following refreshments at 5:30pm. Further networking time with refreshments will follow the discussion. Panel members include:  

• Lucas Nivon, Ph.D. (discussion leader)
Co-founder and CEO, Cyrus Biotechnology 
• Pat Lorton, Ph.D. (panelist)
Chief Technology Officer, Schrodinger
• Essam Metwally, Ph.D. (panelist)
Senior Scientist, Chemical Computing Group
• Craig Bruce, Ph.D. (panelist)
Scientific Software Development Manager, OpenEye Scientific

Molecular Modeling on the Cloud: Implementing complex workflows for all users

Many major advances have been made in the last 10 years in small molecule and protein modeling and design, including the design of novel protein folds and protein-binding activity into inert scaffolds. All of these have required very large amounts of computing and complex workflows with sensitive benchmarking and expert configuration. This complexity, sensitivity and large infrastructure requirements have in many cases limited these tools to very few use cases in an academic environment with access to very large shared compute resources. We propose that the public cloud provides an opportunity to circumvent infrastructure limitations while simultaneously reducing complexity and increasing accessibility.

Here we examine the previously existing infrastructure used to run these types of calculations. We consider the design requirements for a value-focused research organization in a cloud implementation of molecular modeling and design, such as maintenance cost/time, reliability, speed to answers, and correctness of answers. Then we propose various architectural instantiations of a “cloud” modeling suite of software while considering best practices in software-as-a-service tools from other fields (such as email). Finally, we outline one general solution to a cloud architecture that emphasizes these requirements and reduces user complexity. Simplification to users requires shifting complexity to cloud infrastructure and increased software development time.

We focus on one case study of a scientific workflow implementation, detailing architectural/workflow aspects and the scientific benchmarking required.

Special thanks to NIBR for hosting, especially Lewis Whitehead and Erica Gauthier, ACS COMP for sponsoring refreshments

BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, OpenEye Scientific, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Dotmatics, Cresset, Cyrus Biotechnology, Optibrium, Team Arrayo, and Scilligence

Monday, May 1, 2017

BAGIM Event: Drs. Andrew Hopkins and Scott Brown at the Broad Institute in Cambridge, Wednesday May 17th, starting at 6pm.

Dr. Andrew Hopkins of Exscientia Ltd and the University of Dundee and Dr. Scott Brown of Sunovion Pharmaceuticals Inc. will present "Artificial Intelligence approaches to designing psychiatric drugs" on Wednesday, May 17th at the Broad Institute at 415 Main Street in Cambridge, starting at 6pm. Networking and refreshments will follow the talk at Meadhall at 4 Cambridge Center.  Please RSVP for this event.
Abstract:
Investment in drug discovery for psychiatric disease has been in significant decline due to the challenges of translating drugs from the laboratory to the clinic. Our deepening understanding of the genomics of psychiatric diseases holds the promise of better translational to the clinic. However, the price of our deeper understanding is an increase in the complexity of the drug discovery process for new psychiatric drugs. The polygenic model of psychiatric diseases suggests there maybe few stand-alone, individual drug targets. The translational crisis in psychiatric drug discovery requires entirely new ways of thinking about drug discovery. Historically, most treatment options we have today for psychiatric illnesses act on multiple central nervous system (CNS) drug targets, via polypharmacology. Until now successful, multi-target CNS drugs were not designed rationally, but were often discovered by phenotypic screening or clinical observations. Therefore, we need to consider complex ‘target product profiles’ for new medicines that deliberately work by synergy and emergent behavioural properties, by acting on multiple drug targets or by direct phenotypic perturbation. To tackle the complexity in psychiatric drug discovery, Exscientia and Sunovion are collaborating to develop and validate an Artificial Intelligence (A.I.) platform to design drugs rationally against complex ‘target product profiles’ defined by high-content phenotypic and multi-target objectives. We shall describe in this presentation our on-going work on the development of the next generation of psychiatric drugs, that are being designed by algorithm.
BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, OpenEye Scientific, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Dotmatics, Cresset, Cyrus Biotechnology, Optibrium, Team Arrayo, and Scilligence

Friday, April 7, 2017

BAGIM Event: Dr. Todd Martinez, Wednesday May 3rd at Vertex Pharmaceuticals, starting at 5:30pm

Dr. Todd J. Martinez, of Stanford University, will present "Quantum Chemistry and Ab Initio Molecular Dynamics on Graphical Processing Units" on Wednesday, May 3rd, 2017, at Vertex Pharmaceuticals.  Dr. Martinez will present at 7pm, but networking and refreshments will precede the talk starting at 5:30pm.

Please RSVP with your full name to help us plan for the event and to help speed security check in.

Abstract: Solution of the electronic structure problem (often known as “quantum chemistry”) is a critical step in much of chemical simulation. For example, quantum chemistry is required to fit parameters for empirical force fields and also for ab initio molecular dynamics approaches which eschew force fields altogether. The recent introduction of stream processors such as graphical processing units (GPUs) promises to accelerate quantum chemistry calculations. We provide an overview of the differences between GPUs and conventional CPUs and discuss how quantum chemistry algorithms can be redesigned to use GPUs effectively. We show that accelerations of up to three orders of magnitude can be achieved (relative to conventional CPU implementations) by exploiting GPUs with redesigned algorithms. Several applications enabled by these advances are discussed, including an assessment of the size of QM regions needed to achieve convergence in QM/MM methods. Finally, we discuss the ab initio nanoreactor, which uses GPU-enabled ab initio molecular dynamics to discover reaction networks. We present the overall concept of the nanoreactor and some first applications to complex reaction networks in combustion and prebiotic chemistry.

BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, OpenEye Scientific, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Dotmatics, Cresset, Cyrus Biotechnology, Optibrium, Team Arrayo, and Scilligence

Wednesday, March 8, 2017

BAGIM Event: Dr. Meir Glick, Thursday April 13 at Le Meridien in Cambridge, starting at 6pm

Dr. Meir Glick, the Director of Informatics at Merck Research Laboratories, will present "A forward-looking strategy on how can chemistry and informatics enable biology" on Thursday, April 13th, starting at 6pm at Le Méridien Cambridge-MIT, 20 Sidney Street in Cambridge. Food, drinks, and networking will follow at Asgard, 350 Massachusetts Ave in Cambridge.
Please RSVP at to help us plan for this event.
Abstract
This presentation describes how integration of screening, chemical synthesis and informatics can enable target identification and validation. More specifically, design of high quality molecular probes that perturb novel targets and phenotypes, creation of the most biologically relevant assays systems, progressing the right perturbagen into in vivo studies and eventually transforming the organization from data creators into decision makers.
Biography for Meir Glick
Dr. Glick is the Director of Informatics where he is responsible for the vision, strategy and delivery for a world class group that influences all of Merck Research. Before joining Merck in 2015 Dr. Glick was the head of the in silico Lead Discovery at the Novartis Institutes for BioMedical Research where he worked since 2003. He was a trained as a postdoc in the area of computational chemistry in the University of Oxford and received his PhD in computational chemistry from the Hebrew University of Jerusalem. Dr. Glick is the author of over 60 scientific papers in peer reviewed journals, patents and book chapters (https://scholar.google.com/citations?user=fwxyVs8AAAAJ&hl=en)
BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, OpenEye Scientific, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Cresset, Cyrus Biotechnology, Optibrium, Team Arrayo, and Scilligence

Wednesday, January 25, 2017

BAGIM Event: Dr. Matthew Segall Wed March 1 at Millennium Takeda

Dr. Matthew Segall, CEO of Optibrium, will present "Predicting Routes, Sites and Products of Metabolism by Cytochrome P450 " on Wednesday March 1, starting at 6pm at the Millennium Takeda Oncology building at 35 Landsdowne St in Cambridge.  Refreshments and networking will follow at Asgard, 350 Massachusetts Ave in Cambridge.

Please RSVP to help us plan for this event.

Abstract

In this presentation, we will describe recent developments to a method for predicting Cytochrome P450 (P450) metabolism that combines quantum mechanical (QM) simulations to estimate the reactivity of potential sites of metabolism on a compound with a ligand-based approach to account for the effects of orientation and steric constraints due to the binding pockets of different P450 isoforms (DOI: 10.1021/acs.jcim.6b00233). These new developments include modelling reaction pathways for epoxidation and developing models for an extended range of P450 isoforms. The resulting models achieve accuracies of 85-90% on independent test sets.

The metabolites resulting from oxidation at each potential site of metabolism can also be predicted, to guide experimental metabolite identification and highlight potentially active, reactive or toxic metabolites for further investigation. However, as the regioselectivity of metabolism may be different for each isoform, to predict the profile of metabolites that are likely to be formed, it is also necessary to identify which isoforms will be responsible for the metabolism of a compound. We will describe the latest results of our research to develop an accurate ‘WhichP450’ model, which can be coupled with isoform-specific regioselectivity models to generate a ranked list of expected metabolites.

Biography for Matthew Segall

Matt is CEO of Optibrium. He has a Master of Science in computation from the University of Oxford and a Ph.D. in theoretical physics from the University of Cambridge. As Associate Director at Camitro (UK), ArQule Inc. and then Inpharmatica, he led a team developing predictive ADME models and state-of-the-art intuitive decision-support and visualization tools for drug discovery. In January 2006, he became responsible for management of Inpharmatica's ADME business, including experimental ADME services and the StarDrop software platform. Following acquisition of Inpharmatica, Matt became Senior Director responsible for BioFocus DPI's ADMET division and in 2009 led a management buyout of the StarDrop business to found Optibrium, which develops software for small molecule design, optimisation and data analysis. Matt has published over 30 peer-reviewed papers and book chapters on computational chemistry, cheminformatics and drug discovery.

Thursday, January 19, 2017

BAGIM Event: Dr. John Conway & panel discussion on Cloud Computing Tuesday 2/7, at Vertex

Dr. John Conway of LabAnswer will present a lecture concerning cloud computing strategies for discovery research starting at 6pmon Tuesday, February 7th in the 9th floor multipurpose room of building 1 at Vertex Pharmaceuticals at 50 Northern Avenue, Boston. A panel discussion with Q/A on cloud computing will follow with Chris Waller from Merck, Dave Johnson from Moderna, and Steve Litster from Novartis. Networking will follow at approximately 7:30pm, more details TBD. 



Please RSVP  with your real first and last names to speed the sign in process at Vertex. You will find additional information there about the contents of the presentation.


Abstract: Many discovery research organizations struggle with their research informatics environments. Startups and new ventures may have these problems for reasons different than more established discovery research organizations. One observation for startups is the lack of resources or time to focus on strategic thinking and planning because researchers and supporting IT are consumed by tactical priorities and decisions. Defining criteria for decision making and future direction is imperative to a solid scientific data strategy and now with the cloud as an option; an infrastructure strategy. Newer organizations might choose a strict cloud environment where more established organizations will start to migrate towards cloud and will most likely result in a hybrid (on-premise/cloud) environment. From a scientific software vendor perspective, multi-tenant cloud solutions/platforms have the ability to solve and reduce many costly processes, interactions, and problems. From a client perspective, the cloud affords the opportunities of plug and play, reduced internal IT footprint, more efficient collaboration and faster adoption. This BAGIM event is to discuss and generate observations in the discovery research industry with respects to cloud informatics and computational approaches and solutions and to generate a report for the BAGIM community based on the resulting observations and discussions. Topics to be discussed include adoption, implementation, successes, failure, and the future.


BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Cresset, Cyrus Biotechnology, Dotmatics, Optibrium, Team Arrayo, and Scilligence

Friday, January 6, 2017

BAGIM Event: Dr. John Burke Thursday January 19 at The Broad Institute

Dr. John Burke, CEO of Applied BioMath, will present "Quantitative Modeling and Simulation Approaches: Driving Critical Decisions from Research through Clinical Trials" at the Broad Institute at 415 Main Street in Cambridge on Thursday, January 19th.  The talk will begin at 6pm in the auditorium and be followed by refreshments and networking at Meadhall at 4 Cambridge Center starting around 7pm.   Please RSVP for the meeting. 

Abstract: Quantitative Systems Pharmacology (QSP) is a mathematical modeling and engineering approach to translational medicine that aims to quantitatively integrate knowledge about therapeutics with an understanding of its mechanism of action in the context of human disease mechanisms. The goal of QSP modeling is “to understand, in a precise, predictive manner, how drugs modulate cellular networks in space and time and how they impact human pathophysiology [1].” In doing so, QSP approaches de-risk projects, accelerate the development of best in class therapeutics, and reduce late stage attrition rates. This results is helping industry save money, accelerate timelines, and make better therapeutics, ultimately improving patients’ lives. In the five years since the NIH QSP Working Group met last on this topic, progress has been made to advance this science and to integrate QSP approaches in the drug discovery and development process in industry. Here several case studies will be shown that highlights examples of QSP efforts that have accelerated the discovery and development of best-in-class therapeutics, and impacted critical decisions, in the continuum from preclinical exploration to clinical research. Examples include: providing biological understanding, impacts on new target proposals, lead generation, clinical candidate selection, IND support, and clinical trial go/no go decisions from industry.

1. Sorger, PK, et al. Quantitative and Systems Pharmacology in the Post-genomic Era: New Approaches to Discovering Drugs and Understanding Therapeutic Mechanisms. An NIH White Paper, QSP Workshop Group R. Ward, Editor. 2011.

Dr. Burke is CEO, President and Co-founder of Applied BioMath. He earned his PhD in Applied Mathematics from Arizona State University, completed a senior postdoctoral fellowship in Biological Engineering at Massachusetts Institute of Technology, and was Co-Scientific Director of the Cell Decision Process Center at Harvard Medical School, a NIH Center of Excellence. While at MIT and HMS he provided systems consulting support for large and small pharma and biotechs (e.g., AstraZeneca, Pfizer, Momenta), and software companies (e.g., Jacobian and MathWorks). Next he was at Merrimack Pharmaceuticals as a Senior Mathematical Biologist. He then was Global Head of Systems Biology and Pharmacology at Boehringer Ingelheim (BI). He was responsible for starting, implementing, scaling, and managing the worldwide strategy, portfolio, and efforts of BI’s systems initiatives. In a span of five years his group supported over 120 milestone transitions and major decisions from new target stage, lead generation, clinical candidate selection, through clinical trial support. Project indications and platforms included Oncology, Metabolic Disease, CNS, Cardiovascular, Respiratory, Infectious Disease, Inflammation and Immunology for large molecules, fixed dose combinations, ADC, and cell therapies. He is presently an Adjunct Professor at University of Massachusetts, Lowell, developing a QSP course at Harvard Medical School to be taught in the Spring of 2017, an industry advisor for the NIH/NCATS “Translational Center of Tissue Chip Technologies” and the DARPA - MIT Human on a Chip Grant, and an advisor to the Mathematics Department, University of Massachusetts, Lowell


BAGIM is sponsored by Dassault Systemes, DNASTAR, LabAnswer, Schrodinger, Silicon Therapeutics, Acellera, Acpharis, Cambridge Crystallographic Data Centre, Chemical Computing Group, Cresset, Cyrus Biotechnology, Dotmatics, Optibrium, and Scilligence,