BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Developing an optimisation algorithm to supervise active learning 
 in drug discovery - Dr David Marcus\, GSK
DTSTART:20180227T130000Z
DTEND:20180227T140000Z
UID:TALK100807@talks.cam.ac.uk
CONTACT:Dr Vivien Gruar
DESCRIPTION:One of the challenges in drug discovery is to optimise the che
 mical composition of an initial set of promising molecules to improve mult
 iple properties that influence how the drug behaves in the human body\, su
 ch as potency\, bioavailability and possible adverse effects. Computer-aid
 ed drug design facilitate this process by applying machine learning models
  that predict these properties and can virtually consider huge number of p
 ossible molecules to focus the search for molecules that will be tested la
 ter in the lab. However\, some of these models might under-perform with a 
 limited amount of data which may result in suggesting molecules with lower
  activity but high similarity to the initial set. \n\nActive-learning is a
  relatively new approach in drug discovery to assist the selection of pote
 ntial molecules by suggesting those that could also improve these models i
 n an iterative/feedback manner over several cycles and suggest structurall
 y novel molecules. GSK has been implementing this approach that could pote
 ntially reduce costs and development time by the increasing the ability to
  calculate better performing models and suggest molecules with improved mu
 ltiple properties. We are currently evaluating several optimisation algori
 thms to adapt to each iteration and suggest the amount of structurally mod
 ifications with respect to multiple properties needed.\n\nThis well-define
 d project seeks an enthusiastic individual that will investigate which opt
 imisation algorithm performs best by evaluating and comparing their potent
 ial to supervise an active learning process. To do that\, we will also nee
 d to develop the necessary metrics to control these decisions\, preferably
  based on multiple parameter optimisation. The outcome of this short study
  will assist our scientists to evaluate this approach during drug discover
 y processes in the future.\n\n
LOCATION:MR3 Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR