BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Helsinki Institute for Information Technology | HIIT - ECPv4.8.2//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Helsinki Institute for Information Technology | HIIT
X-ORIGINAL-URL:https://www.hiit.fi
X-WR-CALDESC:Events for Helsinki Institute for Information Technology | HIIT
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20191202T141500
DTEND;TZID=Europe/Helsinki:20191202T144500
DTSTAMP:20200122T014740
CREATED:20191126T200018Z
LAST-MODIFIED:20191126T200115Z
UID:8317-1575296100-1575297900@www.hiit.fi
SUMMARY:Bioinformatics afternoon - Edin Husić - Reconstructing perfect phylogenies via branchings in DAGs
DESCRIPTION:Abstract \nThe perfect phylogeny problem and various generalizations of it have been extensively studied in computational biology. We consider the following variant of the problem: Given a binary matrix M\, the Minimum Conflict-Free Row Split problem asks to compute a smallest possible binary matrix M’ which corresponds to a perfect phylogeny and such that each row of M can be obtained as the bitwise OR of rows in M’. We give a new\, more transparent formulation of this NP-hard problem in terms of an optimization problem on the set of branchings in a derived directed acyclic graph. Building on this formulation\, we obtained several new results. Our work relates to chain partitions in partially ordered sets and (classical and weighted) colorings of graphs. \nOriginally\, the problem was proposed for finding the different subtypes of a tumor using DNA sequencing. The equivalent branching formulation of the problem leads to an ILP formulation\, which we implemented for the purpose of the application. Using simulated data\, we showed that it is more accurate than existing methods in reconstructing the original phylogeny. \n \nBio \nEdin is a third year mathematics PhD student at London School of Economics\, under the mentorship of László Végh. He interests broadly in combinatorial optimization and theoretical computer science. More info at: https://zhero9.github.io/. \n
URL:https://www.hiit.fi/event/bioinformatics-afternoon-edin-husic-reconstructing-perfect-phylogenies-via-branchings-in-dags/
LOCATION:University of Helsinki\, Exactum C222\, Kumpula
END:VEVENT
END:VCALENDAR