#32 Deep tensor factorization and a pitfall for machine learning methods with Jacob Schreiber
In this episode we hear from Jacob Schreiber about his algorithm, Avocado.
Avocado uses a neural netwok to factorize a three-dimensional tensor of epigenomic data into the three independent factors corresponding to cell types, assay types, and genomic loci. Avocado can extract a low-dimensional, information-rich summary from the wealth of experimental data from projects like the Roadmap Epigenomics Consortium and ENCODE. Avocado can also predict the data from the experiments that have not been performed.
Jacob also talks about a pitfall he discovered when trying to predict gene expression from a mix of genomic and epigenomic data. As you increase the complexity of a machine learning model, its performance may be increasing for the wrong reason: instead of learning something biologically interesting, your model may simply be memorizing the average gene expression based on the nucleotide sequence.
Links:
Avocado on GitHub Multi-scale deep tensor factorization learns a latent representation of the human epigenome (Jacob Schreiber, Timothy Durham, Jeffrey Bilmes, William Stafford Noble) Completing the ENCODE3 compendium yields accurate imputations across a variety of assays and human biosamples (Jacob Schreiber, Jeffrey Bilmes, William Noble) A pitfall for machine learning methods aiming to predict across cell types (Jacob Schreiber, Ritambhara Singh, Jeffrey Bilmes, William Stafford Noble)
Create your
podcast in
minutes
It is Free