OPI-Struc: Open Protein Instructions for Structures

OPI-Struc is a multimodal instruction-tuning dataset designed for the STELLA project. The dataset was created by BAAI and its related paper was accepted at ACL 2026. The dataset page was last updated on May 12, 2026.

Use Cases

• Instruction-tuning multimodal LLMs based on unified protein sequence-structure encoding.
• Benchmarking protein structure understanding models based on the STELLA project framework.
• Training models for protein functional annotation based on multimodal protein data.

Strengths

• Dataset is specifically designed for a published research project (STELLA) accepted at ACL 2026.
• Dataset is multimodal, likely containing both protein sequence and structural information.

Limitations

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

Provenance

Source

BAAI

Freshness

Last updated 2026-05-12 02:29:45; freshness should be verified.