BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Directed cliques in neuronal networks and persistent homology repr
 esentations - Martina Scolamiero (KTH - Royal Institute of Technology)
DTSTART:20240808T100000Z
DTEND:20240808T110000Z
UID:TALK219304@talks.cam.ac.uk
DESCRIPTION:This talk will be divided in two parts. In the first part we i
 nvestigate the connectivity of directed networks representing brain microc
 ircuits at the level of neurons and synapses. In particular we investigate
  the role of directed cliques within in-silico reconstructions of striatum
 \, the input stage of the basal ganglia. Directed cliques turn out to be i
 nteresting motifs\, to understand changes in the connectivity of striatum 
 when modelling the impact of Parkinson disease progression at the neuron l
 evel [1]. Simplicial structures are the basis for investigating more compl
 ex patterns in weighted networks\, as the ones given by persistent homolog
 y. In the second part of the talk we will discuss stable ranks with learna
 ble parameters as a convenient framework to extract features from persiste
 nt homology that can be used for statistical analysis and machine learning
  [2].\n[1] The impact of Parkinson&rsquo\;s disease on striatal network co
 nnectivity and cortico-striatal drive: an in-silico study\nI Carannante\, 
 M Scolamiero\, JJJ Hjorth\, A Kozlov\, B Bekkouche\, L Guo\, &nbsp\;Arvind
  Kumar\, Wojciech Chach&oacute\;lski\, Jeanette Hellgren Kotaleski. Networ
 k Neuroscience\, 1-35.\n[2]Algebraic Wasserstein distances and stable homo
 logical invariants of data J Agerberg\, A Guidolin\, I Ren\, M Scolamiero&
 nbsp\;- arXiv preprint arXiv:2301.06484\, 2023.
LOCATION:External
END:VEVENT
END:VCALENDAR