BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:  Computational analyses of high-throughput spatial proteomics dat
 a - Dr Laurent Gatto\, Department of Biochemistry\, Cambridge.
DTSTART:20141029T140000Z
DTEND:20141029T150000Z
UID:TALK55322@talks.cam.ac.uk
CONTACT:38916
DESCRIPTION:In biology\, localisation is function. Knowledge of the locali
 sation of proteins within the cell is of paramount importance to assess an
 d study their function. Generation of high quality data is an essential an
 d challenging task\, and multiple research groups have described various e
 fforts and technical procedures to obtain such data. However\, data analys
 is is as critical as data production for insightful biological interpretat
 ion\, and no consistent and robust solutions have been offered to the comm
 unity so far. I will first present classical state-of-the-art data analysi
 s methodologies\, centred around supervised machine learning approaches\, 
 describe their inherent problems and recent breakthroughs that permit new 
 insights into the data. The second part will compare sequence- and experim
 ent-based approaches\, and illustrate how these complementary data sources
  can be utilised to improve mining of the data. Finally\, I will highlight
  some avenues on how computational understanding of the data can be used t
 o optimise the experimental design and thus improve our understanding of t
 he underlying biology.\n
LOCATION:MR4\, Centre for Mathematical Sciences\, Wilberforce Road\, Cambr
 idge
END:VEVENT
END:VCALENDAR