Twadn

Twadn aligns dynamic protein-protein interaction (PPI) networks using time warping to capture temporal variation and improve inference of protein function and phylogenetic relationships.

Key Features:

• Time warping alignment: Applies a time warping strategy to align dynamic PPI networks, capturing temporal variations that are lost in static models.
• Dynamic network focus: Explicitly models dynamic protein-protein interaction (PPI) networks rather than static network snapshots.
• Benchmarking performance: Outperformed existing dynamic network alignment algorithms DynaMAGNA++ and DynaWAVE in comparative evaluations.
• Evaluation metrics: Uses area under the receiver operating characteristic curve (AUC-ROC) and area under the precision-recall curve (AUC-PR) to quantify accuracy.
• Comparison to static methods: Demonstrated advantages over the static network alignment algorithm NetCoffee2 in experiments using the Drosophila protein interaction network.
• Biological inference: Preserves timing information in alignments to support prediction of protein function and analysis of evolutionary relationships.

Scientific Applications:

• Protein function prediction: Improves prediction of protein function by aligning temporal interaction patterns across conditions or species.
• Phylogenetic and evolutionary analysis: Aids elucidation of phylogenetic relationships and molecular-level evolution by incorporating temporal dynamics into network alignments.
• Systems biology temporal analysis: Provides more accurate representations of biological processes over time for systems biology studies.

Methodology:

Employs a time warping strategy to align dynamic networks; comparative evaluations were performed against DynaMAGNA++, DynaWAVE, and the static algorithm NetCoffee2 using AUC-ROC and AUC-PR, including experiments on the Drosophila protein interaction network.