Computational Molecular Design and Data Science
An exciting opportunity is available to apply the latest in data modeling techniques to the realm of drug discovery. The last few decades have seen an unprecedented growth of tools and technology to enable the extraction of relationships within and between complex sets of data. The successful candidate will leverage the tools of drug discovery informatics with the latest in data mining and modeling methods developed in other disciplines and apply them to the prospective modeling of on- and off-target behavior of drug candidates. The primary focus of these roles will be to work closely with collaborators from a variety of groups to provide data-driven decisions/suggestions on potential. In addition, the successful applicant will help to develop and optimize computational approaches that leverage external and proprietary data for the automated design and selection of small molecules from limited knowledge of drug targets. The ability to concisely and accurately communicate complex techniques and the results derived from their application to non-experts is essential.
• Understanding relevant datasets, both those generated internally and those from the public domain.
• Applying chem-informatics methods to support drug discovery, optimization, and development in medicinal chemistry project team settings.
• Applying machine learning algorithms to drug discovery problems and educating colleagues about machine learning approaches.
• Communicating data analysis and modeling results to the collaborators in the GSK scientific community.
• Engaging in scientific programming and algorithm optimization to enhance our software infrastructure.
• MS or PhD in Cheminformatics, Computational Chemistry, Computer Science, Physics or other numerate discipline.
• Expertise in the design, implementation and execution of algorithmic solutions for the collection, curation, analysis, mining and modeling of complex chemical and biological data.
• A scientist who has experience working with small molecule structures and their featurization to answer scientific questions.
• Expertise in one or more programming languages (e.g. Python, Perl, C/C++, java, or R) and chemistry toolkits (e.g. RDKit, OpenEye or Chemaxon)
• Competent working in a Linux/Unix environment
• Demonstrated ability to work in multi-disciplinary teams, displaying excellent interpersonal skills
• Good organizational and communication skills, with the ability to liaise with scientists and external collaborators at all levels across the wider drug discovery organization
• Ability to independently review and appraise scientific literature
•• Solid foundation in chemistry, physics, computer science, statistics, probability theory and analytics.
• Experience with machine learning (supervised and unsupervised), active learning, statistical design of experiments, multi-objective modeling and QSAR techniques.
• Demonstrated expertise in analyzing multiplexed phenotypic or high content screening data.
• Knowledge of drug discovery: medicinal chemistry, toxicology, DMPK.
• Familiarity with workflow/pipeline-based programming tools, databases, high performance, and GPU computing.
• A candidate with experience working and leading on project teams designing of new medicines could see an expansion of responsibilities.