Shypd.aiThe largest open index of AI toolsProteinNet is an open-source dataset for machine learning of protein structure, providing standardized sequences, structures, and alignments. It offers training, validation, and test splits based on CASP assessments.
Trending
About
ProteinNet is a standardized, open-source dataset designed to facilitate machine learning research in protein structure prediction. It compiles protein sequences, their corresponding structures (secondary and tertiary), multiple sequence alignments (MSAs), and position-specific scoring matrices (PSSMs). A key feature is its provision of standardized training, validation, and test splits, which are built upon the biennial CASP assessments. This approach ensures that test sets push the boundaries of computational methodology, mirroring real-world challenges. ProteinNet covers CASP 7 through 12, offering data sets of varying sizes to assess methods across different data richness regimes. It aims to lower the barrier to entry for non-domain experts in machine learning and standardize comparisons across different protein structure prediction methods.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
