Predicting the structural impact of human alternative splicing

Protein structure prediction with neural networks is a powerful new method for linking protein sequence, structure, and function, but structures have generally been predicted for only a single isoform of each gene, neglecting splice variants. To investigate the structural implications of alternative splicing, we use AlphaFold2 to predict the structures of more than 11,000 human isoforms. We employ multiple metrics to identify splicing-induced structural alterations, including template matching score, secondary structure composition, surface charge distribution, radius of gyration, accessibility of post-translational modification sites, and structure-based function prediction.

We identify examples of how alternative splicing induces clear changes in each of these properties. Structural similarity between isoforms largely correlates with degree of sequence identity, but we identify a subset of isoforms with low structural similarity despite high sequence similarity. Exon skipping and alternative last exons tend to increase the surface charge and radius of gyration. Splicing also buries or exposes numerous post-translational modification sites, most notably among the isoforms of BAX. Functional prediction identifies numerous functional differences between isoforms of the same gene, with loss of function compared to the reference predominating. Finally, we use single-cell RNA-seq data from the Tabula Sapiens to determine the cell types in which each structure is expressed.

Our work represents an important resource for studying the structure and function of splice isoforms across the cell types of the human body.