BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Finding Target-Disease-Drug evidences from Literature and supporti
 ng curators. Please note: the details of this talk have changed - Matthew 
 Jeffryes and Santosh Tirunagari\, EMBL-EBI
DTSTART:20230315T140000Z
DTEND:20230315T150000Z
UID:TALK192056@talks.cam.ac.uk
CONTACT:Samantha Noel
DESCRIPTION:Europe PMC is a repository of life sciences research\, includi
 ng peer reviewed research articles and preprints - all freely available fo
 r use via the website (https://europepmc.org). The repository contains ove
 r 42 million abstracts and 8.6 million full text research articles. Assist
 ing researchers with the difficult task of trying to "keep up with the lit
 erature" as output has grown year on year has become a priority in the mac
 hine learning community. As a result\, the important area of “scientific
  services” as a foundation for modern biomedical science is increasing a
 nd the associated use of computation to service this. \nThis talk covers h
 ow machine learning can be applied directly to the literature but also how
  it can be used to assist people whose role it is to curate the bioinforma
 tic content.\nWe present the Europe PMC’s framework for extracting evide
 nces from the scientific research articles. These evidences are currently 
 being submitted to Open Targets (https://opentargets.org)\, a target valid
 ation platform that integrates various evidences to aid drug target identi
 fication and validation. The Europe PMC’s framework harnesses the deep l
 earning methodologies to identify Gene/Protein\, Disease and Chemical/Drug
  associations from the scientific articles and it ranks the articles based
  on their confidence from the Europe PMC literature database.\nWe also pre
 sent the biocuration toolkit\, a combination of machine learning with soft
 ware engineering to develop applications to support biocuration workflows.
  Bioinformatics relies on the provision of databases like UniProt\, PDB\, 
 InterPro\, IntAct and many others. To varying extents\, these databases re
 ly on “curation”: the manual integration of information into a structu
 red format. The curators who perform this task are domain experts\, extrac
 ting information from the biomedical literature and other sources and maki
 ng it accessible to computational methods. Identifying relevant informatio
 n from the literature becomes more difficult as the rate of growth for the
  literature increases year on year. This tool  will assist biocurators wit
 h this process to increase their biocuration capabilities.\n\n
LOCATION:CMS\, Meeting Room 15
END:VEVENT
END:VCALENDAR