ASpdb: an integrative knowledgebase of human protein isoforms from experimental and AI-predicted structures

Abstract

Alternative splicing is a crucial cellular process in eukaryotes, enabling the generation of multiple protein isoforms with diverse functions from a single gene. To better understand the impact of alternative splicing on protein structures, protein-protein interaction and human diseases, we developed ASpdb (https://biodataai.uth.edu/ASpdb/), a comprehensive database integrating experimentally determined structures and AlphaFold 2-predicted models for human protein isoforms. ASpdb includes over 3400 canonical isoforms, each represented by both experimentally resolved and predicted structures, and >7200 alternative isoforms with AlphaFold 2 predictions. In addition to detailed splicing events, 3D structures, sequence variations and functional annotations, ASpdb uniquely offers comparative analyses and visualization of structural alterations among isoforms. This resource is invaluable for advancing research in alternative splicing, structural biology and disease mechanisms.

Graphical Abstract

Figure 1.

Overview of the ASpdb database. (A) Summary of gene isoform structure, gene expression and AS information for 3490 human genes. The left bar plot displays the distribution of AS region lengths in canonical isoforms, while the right bar plot illustrates the distribution of the number of AS isoforms per gene. The information provided for gene summary includes: gene overview, gene isoform structure and gene expression across TCGA and GTEx. The information provided for AS summary include AS knowledge, isoform amino acid sequences, MSAs and functional domain retention. (B) Experimentally solved and AF2-predicted 3D structures of protein isoforms, including active sites and reliability using pLDDT distribution, MSA heatmap and Ramachandran plot. CM: Cryo-electron microscopy. (C) Comparison of 3D structures between canonical and alternative isoforms using TM-scores and statistical tests on changes in structure features. RAS: Relative accessible surface. (D) Overview of protein–protein interactions, clinically important variants, related drugs and associated human diseases.

Figure 2.

An illustration of the comprehensive analysis results of NF2 gene. The detailed information for each panel can be found in Supplementary Tables S1–S5 and Supplementary Figures S1–S2. (A) Gene summary, including gene name, ID and description. (B) Gene ontology terms with evidence from Entrez and associated PubMed IDs. (C) AS and isoform information for three NF2 isoforms, listing canonical and alternative spliced isoforms and modifications. (D) Conversion tables of UniProt, Ensembl and RefSeq IDs, specific to the three isoforms. (E) Gene structures of canonical and alternatively spliced genes, along with their expression levels across GTEx tissues, visualized as heatmaps.

Figure 3.

Detailed structural information of APH-1 gene and its isoforms. (A) AF2-predicted structure of the long isoform APH-1aL (Q96BI3-1) and its pLDDT distribution. (B) AF2-predicted structure of the short isoform APH-1aS (Q96BI3-2) and its pLDDT distribution. (C) Superimposed structure of APH-1aL (Q96BI3-1) and APH-1aS (Q96BI3-2) with the corresponding TM-score. (D) Comparison of secondary structure and RAS area changes between APH-1aL (Q96BI3-1) and APH-1aS (Q96BI3-2). (E) Drugs targeting APH1A isoforms.