AI-Powered Enzyme Engineering Data for Sitagliptin Synthesis

1.39 million structural fragments inform an AI-driven protein-engineering framework for biocatalyst development. The dataset, authored by Pravin Kumar R and last updated in May 2026, details the prediction and validation of productive enzyme substitutions. Five AI-prioritized enzyme variants, each containing nine mutations, were evaluated and scaled to 7 L fermentation.

Use Cases

• Training AI models for enzyme property prediction based on interaction energy descriptors and solvent effects.
• Benchmarking biocatalyst performance based on conversion rates and enantiomeric excess metrics.
• Optimizing sustainable chemical processes based on data comparing DMSO and biodegradable cosolvent systems.

Strengths

• Dataset size is 1.1 MB, indicating a focused and manageable download.
• Description provides specific performance metrics, including 51% conversion at 20% enzyme load and >99% enantiomeric excess.
• The AI framework was validated at industrial scale, with variants tested in 7 L fermentation.

Limitations

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

Provenance

Source

figshare

Collection Method

Likely contains computational predictions and experimental validation data from an AI-driven 6D-grid protein-engineering framework.

Freshness

Last updated 2026-05-14 12:34:34; freshness should be verified.