DeepFold: enhancing protein structure prediction through optimized loss functions, improved template features, and re-optimized energy function.

MOTIVATION: Predicting protein structures with high accuracy is a critical challenge for the broad community of life sciences and industry. Despite progress made by deep neural networks like AlphaFold2, there is a need for further improvements in the quality of detailed structures, such as side-chains, along with protein backbone structures. RESULTS: Building upon the successes of AlphaFold2, the modifications we made include changing the losses of side-chain torsion angles and frame aligned point error, adding loss functions for side chain confidence and secondary structure prediction, and replacing template feature generation with a new alignment method based on conditional random fields. We also performed re-optimization by conformational space annealing using a molecular mechanics energy function which integrates the potential energies obtained from distogram and side-chain prediction. In the CASP15 blind test for single protein and domain modeling (109 domains), DeepFold ranked fourth among 132 groups with improvements in the details of the structure in terms of backbone, side-chain, and Molprobity. In terms of protein backbone accuracy, DeepFold achieved a median GDT-TS score of 88.64 compared with 85.88 of AlphaFold2. For TBM-easy/hard targets, DeepFold ranked at the top based on Z-scores for GDT-TS. This shows its practical value to the structural biology community, which demands highly accurate structures. In addition, a thorough analysis of 55 domains from 39 targets with publicly available structures indicates that DeepFold shows superior side-chain accuracy and Molprobity scores among the top-performing groups. AVAILABILITY AND IMPLEMENTATION: DeepFold tools are open-source software available at https://github.com/newtonjoo/deepfold.

Quality of life in intermittent exotropia for Korean children and their parents.

BACKGROUND: Patients with strabismus are more likely to have mental health problems, including high rates of depressive symptoms and social phobia. Intermittent exotropia (IXT) typically occurs in early childhood and is more common in Asian populations. We aim to assess the health-related quality of life (HRQOL) concerns in children with intermittent exotropia (IXT) using the Intermittent Exotropia Questionaire (IXTQ), and their associations with the clinical severity of IXT and the parents' HRQOL concerns. METHODS: IXT, defined as both distance and near exodeviation ≥ 10 prism diopters were eligible for inclusion. The final IXTQ score is calculated using the mean score for all items, and ranges from 0 (worst HRQOL) to 100 (best HRQOL). The correlations of child IXTQ scores with their deviation angle and stereoacuity were measured, as were those with their parent's IXTQ scores. RESULTS: One hundred twenty-two children with IXT (aged 5-17 years) and one parent for each child completed the child and parent IXTQ, respectively. The greatest HRQOL concern for each child with IXT and their parent was "Worry about eyes" (frequency 88%, score 35.0 ± 27.8). Lower child IXTQ scores were associated with a larger distance (r = 0.24, p = 0.007) and near deviation angle (r = 0.2, p = 0.026). "It bothers me because I have to wait for my eyes to clear up" and "Waiting for their eyes to clear up" were more common in children with a larger deviation angle (both p < 0.05). The parent IXTQ scores (52.1 ± 25.3) were lower than the child ones (79.7 ± 15.8) and showed a positive correlation with child IXTQ scores (r = 0.26, p = 0.004). Lower parent IXTQ scores were associated with poor distance stereoacuity (r = 0.23, p = 0.01). CONCLUSION: The HRQOL of IXT children was positively related to that of their parents. A larger deviation angle and worse distance stereoacuity function may predict more-negative impacts on children and their parents, respectively.