SOTAVerified

Protein Design

Formally, given the design requirements of users, models are required to generate protein amino acid sequences that align with those requirements.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 21–30 of 175 papers

TitleStatusHype
Out of Many, One: Designing and Scaffolding Proteins at the Scale of the Structural Universe with Genie 2Code2
RNAFlow: RNA Structure & Sequence Design via Inverse Folding-Based Flow MatchingCode2
Generating Novel, Designable, and Diverse Protein Structures by Equivariantly Diffusing Oriented Residue CloudsCode1
Improving large language models with concept-aware fine-tuningCode1
Iterative Refinement Graph Neural Network for Antibody Sequence-Structure Co-designCode1
Fast non-autoregressive inverse folding with discrete diffusionCode1
Generative De Novo Protein Design with Global ContextCode1
Fold2Seq: A Joint Sequence(1D)-Fold(3D) Embedding-based Generative Model for Protein DesignCode1
Improving few-shot learning-based protein engineering with evolutionary samplingCode1
Bridge-IF: Learning Inverse Protein Folding with Markov BridgesCode1
Show:102550
← PrevPage 3 of 18Next →
All datasetsCATH 4.2CATH 4.3

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1GraphTransPerplexity6.63—Unverified
2StructGNNPerplexity6.4—Unverified
3AlphaDesignPerplexity6.3—Unverified
4GCAPerplexity6.05—Unverified
5GVPPerplexity5.36—Unverified
6ProteinMPNNPerplexity4.61—Unverified
7PiFoldPerplexity4.55—Unverified
8Knowledge-DesignPerplexity3.46—Unverified
#ModelMetricClaimedVerifiedStatus
1ESM-IFPerplexity6.44—Unverified
2GVP-largePerplexity6.17—Unverified