Accurate de novo protein structure prediction remains a fundamental challenge, particularly in cases where homologous templates are unavailable or evolutionary information is weak. While end-to-end methods such as AlphaFold2 have achieved unprecedented accuracy, their closed box nature provides limited insight into the folding process and offers little flexibility for incorporating external evaluation. Here, we investigate whether model quality assessment (MQA) can be integrated into the structure prediction pipeline as a closed-loop feedback mechanism to iteratively improve prediction accuracy. In this study, we propose DGMFold, a de novo protein structure prediction method that establishes a feedback loop among three components: the geometric constraint prediction network (GeomNet), the structural simulation module, and the model quality evaluation network (EmaNet). In GeomNet, co-evolutionary features extracted from multiple sequence alignments (MSAs) are fed into an improved residual neural network to predict inter-residue geometric constraints, which are then used to guide structure folding. EmaNet then extracts 1D and 2D features from the folded structure model and employs a deep residual neural network to estimate the inter-residue distance deviation and per-residue lDDT. These evaluations are subsequently fed back into GeomNet as dynamic features, enabling iterative refinement of the predicted geometries and overall model accuracy. DGMFold was tested on 437 benchmark proteins and 20 FM targets of CASP14. Experimental results demonstrate that the closed-loop feedback mechanism significantly contributes to the performance of DGMFold, and the prediction accuracy of DGMFold outperforms that of the state-of-the-art de novo methods trRosetta and RaptorX at the time. When evaluated on the 124 human proteins for which AlphaFold2 yields TM-scores below 0.9, DGMFold achieves higher prediction accuracy than AlphaFold2 and RoseTTAFold on 71 and 72 targets, respectively, and outperforms both on 58 proteins.

The scoring models used for protein structure modeling and ranking are mainly divided into unified field and protein-specific scoring functions. Although protein structure prediction has made tremendous progress since CASP14, the modeling accuracy still cannot meet the requirements to a certain extent. Especially, accurate modeling of multi-domain and orphan proteins remains a challenge. Therefore, an accurate and efficient protein scoring model should be developed urgently to guide the protein structure folding or ranking through deep learning. In this work, we propose a protein structure global scoring model based on equivariant graph neural network (EGNN), named GraphGPSM, to guide protein structure modeling and ranking. We construct an EGNN architecture, and a message passing mechanism is designed to update and transmit information between nodes and edges of the graph. Finally, the global score of the protein model is output through a multilayer perceptron. Residue-level ultrafast shape recognition is used to describe the relationship between residues and the overall structure topology, and distance and direction encoded by Gaussian radial basis functions are designed to represent the overall topology of the protein backbone. These two features are combined with Rosetta energy terms, backbone dihedral angles and inter-residue distance and orientations to represent the protein model and embedded into the nodes and edges of the graph neural network. The experimental results on the CASP13, CASP14 and CAMEO test sets show that the scores of our developed GraphGPSM have a strong correlation with the TM-score of the models, which are significantly better than those of the unified field score function REF2015 and the state-of-the-art local lDDT-based scoring models ModFOLD8, ProQ3D and DeepAccNet, etc. The modeling experimental results on 484 test proteins demonstrate that GraphGPSM can greatly improve the modeling accuracy. GraphGPSM is further used to model 35 orphan proteins and 57 multi-domain proteins. The results show that the average TM-score of the models predicted by GraphGPSM is 13.2 and 7.1% higher than that of the models predicted by AlphaFold2. GraphGPSM also participates in CASP15 and achieves competitive performance in global accuracy estimation.

MOTIVATION: With the great progress of deep learning-based inter-residue contact/distance prediction, the discrete space formed by fragment assembly cannot satisfy the distance constraint well. Thus, the optimal solution of the continuous space may not be achieved. Designing an effective closed-loop continuous dihedral angle optimization strategy that complements the discrete fragment assembly is crucial to improve the performance of the distance-assisted fragment assembly method. RESULTS: In this article, we proposed a de novo protein structure prediction method called IPTDFold based on closed-loop iterative partition sampling, topology adjustment and residue-level distance deviation optimization. First, local dihedral angle crossover and mutation operators are designed to explore the conformational space extensively and achieve information exchange between the conformations in the population. Then, the dihedral angle rotation model of loop region with partial inter-residue distance constraints is constructed, and the rotation angle satisfying the constraints is obtained by differential evolution algorithm, so as to adjust the spatial position relationship between the secondary structures. Finally, the residue distance deviation is evaluated according to the difference between the conformation and the predicted distance, and the dihedral angle of the residue is optimized with biased probability. The final model is generated by iterating the above three steps. IPTDFold is tested on 462 benchmark proteins, 24 FM targets of CASP13 and 20 FM targets of CASP14. Results show that IPTDFold is significantly superior to the distance-assisted fragment assembly method Rosetta_D (Rosetta with distance). In particular, the prediction accuracy of IPTDFold does not decrease as the length of the protein increases. When using the same FastRelax protocol, the prediction accuracy of IPTDFold is significantly superior to that of trRosetta without orientation constraints, and is equivalent to that of the full version of trRosetta. AVAILABILITYAND IMPLEMENTATION: The source code and executable are freely available at https://github.com/iobio-zjut/IPTDFold. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.