Data: 14 e 15 de fevereiro
Hora: Das 10h às 11h
Local: Online (via ZOOM)
«O projeto OLISSIPO – Fostering Computational Biology Research and Innovation in Lisbon – anuncia as próximas palestras nos dias 14 e 15 de fevereiro, com os oradores Prof. Jonas Almeida (National Cancer Institute) e Prof. Dan Gusfield (University of California), respetivamente.
Dia 14 de fevereiro
Prof. Jonas Almeida
Título: “Universal Sequence Maps, from biological sequences to numbers and back”
Resumo: Universal Sequence Maps (USM) are an iterated mapping technique that can represent sequence succession in a manner that is both bijective and scale-free. In its original unidirectional form, proposed in 1990 as Chaos Game Representation (CGR), it was proposed as a graphical representation of gene structure that borrowed methodologically from statistical mechanics of complex dynamic phenomena. Over a decade later, CGR was found to provide an alignment-free route to sequence analysis that generalizes Markovian succession. In the subsequent 2 decades, new properties were uncovered and the iterated technique was reformulated, as reviewed in this presentation. Intriguingly, the modern bidirectionally iterated form, USM, appears to offer a convenient route to develop language models of Biological sequences for generative AI applications.
Breve Biografia: I hold the position of (tenured) Senior investigator and Data Science director at the National Cancer Institute, Division of Cancer Epidemiology and Genetics (NIH/NCI intramural program). I also maintain academic positions at the State University of New York (SUNY – Stony Brook), Queen’s University Belfast, UK, and George Mason University in Virginia. I started my academic career at the University of Lisbon (PhD Biological Engineering, 1995), where I remain involved in collaborative initiatives towards Artificial intelligence, Cloud Computing, and the development of consumer-facing polygenic risk calculators for Precision Prevention.
Dia 15 de fevereiro
Prof. Dan Gusfield
Título: “Integer Linear Programming and SAT-Solving in Computational Biology”
Resumo: In this talk I will explain how to use Integer Linear Programming for problems in computational biology, focussing on the problem of RNA folding. This approach has been very successful for a wide range of problems in computational biology. In writing a book on integer programming in computational biology (2019), I examined in depth about 50 such problems. Most solved successfully with ILP. However, there were a few that did not. Then students and I tried using SAT-solving on those harder problems and found that it was more successful on those problems than was ILP, giving another computational tool for computational biology that complements ILP.
Breve Biografia: Prof. Emeritus Dan Gusfield started work in computational biology (around 1982) before the field had a name. He is a Fellow of the ISCB and ACM and was the founding Editor-in-chief of the IEEE/ACM Transactions on Computational Biology and Bioinformatics, and wrote three books on topics in computational biology (plus two others including one that appeared just last week: Proven Impossible). He wrote the first paper indexed in PubMed with the term “computational biology”. He worked (part-time) at the first DOE Human Genome Center (before NIH became involved in the HGP) and was on the three-person (animal, vegetable, mineral) committee that wrote the proposal to establish the Genome Center at UC Davis, resulting in the largest building on the UCD campus, and hiring of multiple new faculty at UC Davis.
