Timezone: »

 
Oral
Protein contact prediction from amino acid co-evolution using convolutional networks for graph-valued images
Vladimir Golkov · Marcin Skwark · Antonij Golkov · Alexey Dosovitskiy · Thomas Brox · Jens Meiler · Daniel Cremers

Wed Dec 07 03:00 AM -- 03:20 AM (PST) @ Area 1 + 2
in Applications »

Proteins are the "building blocks of life", the most abundant organic molecules, and the central focus of most areas of biomedicine. Protein structure is strongly related to protein function, thus structure prediction is a crucial task on the way to solve many biological questions. A contact map is a compact representation of the three-dimensional structure of a protein via the pairwise contacts between the amino acid constituting the protein. We use a convolutional network to calculate protein contact maps from inferred statistical coupling between positions in the protein sequence. The input to the network has an image-like structure amenable to convolutions, but every "pixel" instead of color channels contains a bipartite undirected edge-weighted graph. We propose several methods for treating such "graph-valued images" in a convolutional network. The proposed method outperforms state-of-the-art methods by a large margin. It also allows for a great flexibility with regard to the input data, which makes it useful for studying a wide range of problems.

Keywords: (Application) Bioinformatics and Systems Biology
[ Video] 

Author Information

Vladimir Golkov (Technical University of Munich)
Marcin Skwark (Vanderbilt University)

Dr Marcin J. Skwark is the Bioinformatics Lead at InstaDeep Ltd. Before joining InstaDeep in the beginning of 2020, he conducted academic research in computational structural biology at universities in Stockholm (Stockholm University), Helsinki (Aalto), Beijing (Tsinghua), Nashville (Vanderbilt), and Cambridge, UK (University of Cambridge). He researches on application of machine learning to modelling complex biological systems, structural bioinformatics (esp. protein design), as well as computational methods to enable groundbreaking discoveries in life sciences.

Antonij Golkov (University of Augsburg)
Alexey Dosovitskiy (University of Freiburg)
Thomas Brox (University of Freiburg)
Jens Meiler (Vanderbilt University)
Daniel Cremers (Technical University of Munich)

More from the Same Authors