AI-Prioritized Enzyme Variants for Sitagliptin Synthesis

Pravin Kumar R published a dataset on figshare in May 2026 detailing an AI-driven protein-engineering framework. The dataset likely contains structural fragment data and performance metrics for engineered enzyme variants. It describes a method that used a dataset of 1.39 million structural fragments to predict substitutions, resulting in five variants achieving high conversion rates for the drug sitagliptin.

Use Cases

• Training AI models for protein function prediction based on structural fragment data.
• Benchmarking enzyme engineering methods based on solubility and catalytic stability metrics.
• Optimizing green chemistry processes based on solvent system performance data.
• Analyzing the relationship between mutation combinations and industrial-scale biocatalyst efficiency.

Strengths

• Describes a specific, validated engineering outcome with five variants achieving >99% enantiomeric excess.
• Includes performance metrics at a 7 L fermentation scale, indicating industrial relevance.
• Integrates a large reference dataset of 1.39 million structural fragments.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small (18.1 KB), indicating limited scope.

Provenance

Source

Pravin Kumar R via figshare.

Collection Method

Likely generated from computational protein design and experimental validation.

Freshness

Last updated 2026-05-14 12:34:38; freshness should be verified.