BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Statistics
SUMMARY:Principal Nested Shape Space Analysis of Molecular
Dynamics Data - Ian Dryden (Nottingham)
DTSTART;TZID=Europe/London:20161118T160000
DTEND;TZID=Europe/London:20161118T170000
UID:TALK67490AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/67490
DESCRIPTION:Molecular dynamics simulations produce huge datase
ts of temporal sequences of molecules. It is of in
terest to summarize the shape evolution of the mol
ecules in a succinct\, low-dimensional representat
ion. However\, Euclidean techniques such as princi
pal components analysis (PCA) can be problematic a
s the data may lie on a manifold which is far from
being flat. Principal nested spheres involves the
backwards fitting of a sequence of nested spheres
to data\, and can lead to striking insights which
may be missed using PCA (Jung\, Dryden and Marron
\, 2012\, Biometrika). We develop principal nested
shape spaces (PNSS) for three-dimensional shape d
ata\, and provide some fast fitting algorithms. Th
e methodology is applied to a large set of 100 run
s of 3D protein simulations\, investigating bioche
mical function in applications in Pharmaceutical S
ciences. The data exhibit distinct clusters\, repr
esenting different molecular states\, and these fe
atures are far more apparent using PNSS compared t
o PCA.\n \nThis is joint work with Huiling Le and
Kwang-Rae Kim (University of Nottingham).
LOCATION:MR12\, Centre for Mathematical Sciences\, Wilberfo
rce Road\, Cambridge.
CONTACT:Quentin Berthet
END:VEVENT
END:VCALENDAR