acorde

acorde constructs isoform co-usage networks by integrating bulk long-read sequencing and single-cell RNA-seq (scRNA-seq) data to identify co-differential isoform usage (coDIU) and predict functional elements involved in alternative splicing (AS) across cell types.

Key Features:

• Integration of Bulk Long Reads and Single-Cell Data: Combines bulk long-read sequencing with scRNA-seq to leverage full-length isoform definitions alongside single-cell expression resolution.
• Percentile Correlations Methodology: Employs percentile correlations to mitigate scRNA-seq sparsity and improve accuracy of isoform co-expression estimates.
• Isoform Co-Expression Network Construction: Clusters correlated isoforms into co-expression networks to reveal cell type-specific patterns of alternative isoform usage.
• Detection and Characterization of Co-Differential Isoform Usage (coDIU): Identifies genes exhibiting co-differential isoform usage by selecting same-gene isoforms from constructed networks.
• Functional Element Prediction: Predicts functional elements, including motifs and domains, within long-read-defined isoforms to infer coordinated post-transcriptional regulation.

Scientific Applications:

• Regulatory role investigation: Map isoform co-usage networks to study how alternative splicing regulates biological processes across cell types.
• Biomarker discovery: Identify isoform-level biomarkers for diseases associated with altered splicing patterns.
• Comparative and evolutionary studies: Compare isoform co-usage patterns across conditions or species to explore evolutionary aspects of gene regulation.

Methodology:

Data preprocessing to integrate bulk long-read and scRNA-seq datasets; application of percentile correlations to derive co-expression estimates; network analysis to cluster co-expressed isoforms; and functional annotation of long-read-defined isoforms.