PPI Prediction Results for Six CASP14 Targets Using AF2-Related Methods

Protein-protein interaction prediction results for six targets from the CASP14 experiment. The data includes results from ZDOCK with AF2 models, OmegaFold-linker, AF2-linker, and two versions of AF2-Multimer. Yu Yamamori authored this dataset, which corresponds to a figure in a 2022 Biophysical Reviews article.

Use Cases

• Benchmarking AlphaFold2-Multimer performance based on results for six CASP14 targets.
• Comparing protein-protein interaction prediction methods such as ZDOCK and OmegaFold-linker.
• Analyzing the impact of model version on prediction results based on AF2-Multimer v2.1.1 and v2.2.0.
• Studying prediction limitations for long protein sequences based on the omission of target T1032 from OmegaFold-linker results.

Strengths

• Results are provided for six specific CASP14 targets (T1032, T1038, T1054, T1078, H1045, H1065).
• Includes predictions from five distinct methods or model configurations, enabling comparative analysis.
• Dataset is directly linked to a published scientific figure, providing academic context.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

paperswithcode

Collection Method

Results from computational experiments using specified prediction methods on CASP14 targets.