Abstract

We will follow up on the presentation from the previous week. Exploring a paper
that successfully applied NLP methods, namely transformer architectures, and
transfer-learning, to various classification problems of protein sequences. We
will mainly focus on the learning methods and objectives, transformers
architecture, and data structure. I will present how this topic relates to the
work we (plan to) do in the CEITEC Bioinformatics lab. We will discuss the
nature of genomic data (DNA, RNA, protein sequences) and what other NLP
techniques could be used on the genomic data.