DATAIA Seminar 09/21 | Laurent JACOB « Learning from biological sequences in functional and evolutionary genomics »
2023-09-21 12:30
2023-09-21 14:00
DATAIA Seminar 09/21 | Laurent JACOB « Learning from biological sequences in functional and evolutionary genomics »
Laurent JACOB (Laboratory of Computational and Quantitative Biology - Sorbonne University, LCQB, UMR 7238) will present his work on "Learning from biological sequences in functional and evolutionary genomics". Laurent Jacob is interested in the development of statistical and machine learning methods to solve problems in molecular biology.
Abstract:
- Microbial GWAS: tools to identify genetic determinants of phenotypic traits such as antimicrobial resistance. The importance of accessory genes in microbes makes the usual SNP approach inappropriate. I am developing solutions that rely on k-mers, i.e. the presence of short sequences in genomes. presence of short sequences in genomes.
- Prediction from biological sequences: neural networks that take a biological sequence as input and predict a property of that sequence. biological sequence as input and predict a property of that sequence. This applies for example to regulatory genomics or to fold prediction. or to fold prediction. I am working on regularization and statistical inference of statistical inference on the features extracted by these networks.
- Machine learning for evolutionary genomics: neural networks to infer parameters of sequence evolution models. parameters of sequence evolution models. For some complex models, likelihood maximization is too difficult but sampling is easy. I use a large amount of simulated data to learn a function that inverts the model, and goes for example from a gene family to a phylogeny.
Laurent JACOB (Laboratory of Computational and Quantitative Biology - Sorbonne University, LCQB, UMR 7238) will present his work on "Learning from biological sequences in functional and evolutionary genomics". Laurent Jacob is interested in the development of statistical and machine learning methods to solve problems in molecular biology.
Abstract:
- Microbial GWAS: tools to identify genetic determinants of phenotypic traits such as antimicrobial resistance. The importance of accessory genes in microbes makes the usual SNP approach inappropriate. I am developing solutions that rely on k-mers, i.e. the presence of short sequences in genomes. presence of short sequences in genomes.
- Prediction from biological sequences: neural networks that take a biological sequence as input and predict a property of that sequence. biological sequence as input and predict a property of that sequence. This applies for example to regulatory genomics or to fold prediction. or to fold prediction. I am working on regularization and statistical inference of statistical inference on the features extracted by these networks.
- Machine learning for evolutionary genomics: neural networks to infer parameters of sequence evolution models. parameters of sequence evolution models. For some complex models, likelihood maximization is too difficult but sampling is easy. I use a large amount of simulated data to learn a function that inverts the model, and goes for example from a gene family to a phylogeny.