Crystallization-Induced Diastereomer Transformation Predictions with Feature Analysis

Sara L. McCormack published a dataset on figshare in 2026 for predicting diastereoselectivity in crystallization-induced diastereomer transformations (CIDTs). The data supports a workflow to featurize product structures and construct statistical models, validated on six previously untested substrates. The 3.0 MB dataset is provided in XLSX format.

Use Cases

• Training machine learning models to predict diastereoselectivity based on structural features.
• Analyzing the influence of amide identity on stereochemical outcomes in CIDT reactions.
• Investigating relationships between conformational compactness and crystallization selectivity.
• Validating statistical models for reaction design using the six provided test substrates.

Strengths

• Dataset is 3.0 MB, indicating a focused and manageable size for analysis.
• Model validation was performed on six previously untested substrates, providing a concrete test case.
• Feature analysis identifies specific determinants like amide identity and conformational compactness.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Sara L. McCormack via figshare

Collection Method

Data likely generated from a computational workflow to featurize product structures and construct statistical models.

Freshness

Last updated 2026-06-04 17:07:31; freshness should be verified.