Protein-protein interactions (PPIs) are central to cellular signaling and regulation, and their dysregulation underlies many diseases. Predicting the impact of mutations on PPI stability, quantified as ΔΔG, is essential for understanding disease mechanisms and guiding protein engineering. Here, we first present MutPPI, a graph-based deep-learning model that encodes full-residue structural features of protein-protein complexes and employs a shared GIN-GAT feature extractor for wild-type and mutant complexes. MutPPI outperforms 12 existing methods on an antibody-antigen single-point mutation dataset (S645). By integrating evolutionary information from protein language models, we further develop MutPPI-plus, achieving enhanced predictive performance. Second, we proposed a mutation-path-based data augmentation strategy, which enriches input modalities and improves generalization of both MutPPI and MutPPI-plus. After data augmentation, MutPPI-plus demonstrates state-of-the-art performance on S645 and three additional multi-point mutation datasets (SM_ZEMu, SM595, SM1124), substantially surpassing DDMut-PPI. Our analyses highlight the benefits of the multimodal framework and the physically informed data augmentation method. Together, these results provide a versatile computational tool for accurate ΔΔG prediction, advancing rational protein design.

This study unveils a cutting-edge camera-based system for the automated monitoring of hand hygiene practices within healthcare settings. Utilizing advanced computer vision and machine learning technologies, our system employs three strategically placed synchronized cameras around a wash basin. These cameras capture the handwashing process from multiple perspectives, allowing for detailed analysis of hand movements including finger and wrist dynamics. The extracted skeletal coordinate data are processed by a Gesture Category Model (GCM), which automatically identifies handwashing gestures. The model is rigorously trained on a dataset comprising video recordings from 55 healthcare professionals, focusing on the World Health Organizations seven-step hand-washing protocol. Furthermore, we introduce a Counting Algorithm to quantify the frequency and duration of each gesture, coupled with a Quality Assessment Model (QAM) that evaluates compliance with hand hygiene standards. The systems precision and its strong correlation with expert annotations highlight its potential to significantly enhance hand hygiene compliance and reduce healthcare-associated infections.

CD8+ T cells can recognize peptides presented by class I human leukocyte antigen (HLA-I) of nucleated cells. Exploring this immune mechanism is essential for identifying T-cell vaccine targets in cancer immunotherapy. Over the past decade, the wealth of data generated by experiments has spawned many computational approaches for predicting HLA-I binding, antigen presentation and T-cell immune responses. Nevertheless, existing HLA-I binding and antigen presentation prediction approaches suffer from low precision due to the absence of T-cell receptor (TCR) recognition. Direct modeling of T-cell immune responses is less effective as TCR recognition's mechanism still remains underexplored. Therefore, directly applying these existing methods to screen cancer neoantigens is still challenging. Here, we propose a novel immune epitope prediction method termed IEPAPI by effectively incorporating antigen presentation and immunogenicity. First, IEPAPI employs a transformer-based feature extraction block to acquire representations of peptides and HLA-I proteins. Second, IEPAPI integrates the prediction of antigen presentation prediction into the input of immunogenicity prediction branch to simulate the connection between the biological processes in the T-cell immune response. Quantitative comparison results on an independent antigen presentation test dataset exhibit that IEPAPI outperformed the current state-of-the-art approaches NetMHCpan4.1 and mhcflurry2.0 on 100 (25/25) and 76% (19/25) of the HLA subtypes, respectively. Furthermore, IEPAPI demonstrates the best precision on two independent neoantigen datasets when compared with existing approaches, suggesting that IEPAPI provides a vital tool for T-cell vaccine design.

Given the severe impact of low bone density (LBD) on public health, and to avoid the potential damage of X-rays-based bone density measurements, this study aimed to develop a scoring system for the mass screening for LBD in women aged 50 years or older, using the basic body check-up items as variables. Five variables, including age, body mass index (BMI), systolic blood pressure (SBP), blood glucose level, and total cholesterol level (TCL), were obtained from medical check-up records of 1525 women aged 50 years or older who had done body examination between 2011 and 2018, and were used to construct a scoring system for the screening for LBD. Multivariate logistic regression was applied to investigate the putative association of the five variables with LBD. A scoring system was derived from the regression model to discriminate persons at risk of LBD from low-risk persons. An artificial neural network (ANN) model was also applied to the same task. Precision, recall,$F1$-score, and c-statistic were adopted as evaluation metrics. Age, BMI, SBP, glucose, and TCL were significantly associated with the risk of LBD. Precision, recall, c-statistic, and$F1$-score of the proposed scoring system were 0.66, 0.83, 0.73, and 0.74, respectively. ANNs achieved better performances in terms of all measurements. This study demonstrates the feasibility of using routine body check-up items to estimate LBD risk. Different from X-rays-based instruments, the scoring system derived from this study may serve as a postcheck-up mass screening tool to enable health practitioners to identify individuals at a risk of LBD efficiently and nonintrusively.

Recent studies show that deep learning models perform well in many medical tasks such as medical imaging and automated diagnosis. With qualified training datasets, some models can achieve or even surpass expert-level performance on some tasks. However, as a typical black-box-style approach, deep learning lacks theoretical interpretability, which is especially important for medical tasks. On the other hand, there are many sources of domain knowledge for medical diagnosis from human experts, such as clinical guidelines. How to sufficiently integrate human knowledge in the model is crucial for explainable diagnosis. In this paper, we propose a novel framework for explainable automated diagnosis that leverages explicit medical knowledge. We automate the knowledge extraction from textual clinical guidelines with prompt-based learning, train a set of weighted first-order logical rules with constructed evidence database, and finally infer the diagnosis result with integrated knowledge and multi-sourced data. We instantiate the framework for pulmonary disease diagnosis, and our experiments on a real dataset show that our method outperforms the state-of-the-art baselines in accuracy and interpretability.

In a community question-answering (CQA) system, the answer selection task is used to identify the best answer for a specific question. This plays a key role in improving service quality by recommending appropriate answers to new questions. Recent advances in CQA answer selection have focused on enhancing performance by incorporating community information, and particularly the expertise (previous answers) and authority (position in the social network) of a respondent. However, existing approaches to incorporating this information are limited, as they (a) consider either the expertise or the authority, but not both; (b) ignore domain knowledge that could differentiate between the topics of previous answers; or (c) simply use authority information to adjust the similarity score, rather than fully integrating it into the process of measuring the similarity between the question and answer segments. We propose a new approach called the knowledge-enhanced attentive answer selection (KAAS) model, which enhances performance by (a) considering both the expertise and the authority of the answerer; (b) utilizing human-labeled tags, a taxonomy of tags, and votes as domain knowledge to infer the expertise of the respondent; (c) using a matrix decomposition of the social network (based on ‘following’ relationships) to infer the authority of the respondent and incorporating this information into the process of evaluating the similarity between segments. In addition, we incorporate an external knowledge graph to capture more professional information for CQA systems for vertical communities. We also adopt an attention mechanism to integrate our analysis of both questions and answers texts and the aforementioned community information. Experiments with both vertical and general CQA sites demonstrate the superior performance of the proposed KAAS model.

Different estimators have different optimization criteria according to the concrete application considered. Most existing metrics on estimation performance are some averages of estimation error terms, which usually give “big” or “small” results to show the “bad” or “good” performance of the evaluated estimators. These kinds of metrics are only insufficient statistics of discrete set of estimation errors in some sense and reflect certain narrow aspects of estimation performance. However, an error distribution function is important information which is usually overlooked. To handle this problem, a metric, called desirability level, is provided in [1] to measure how the probability density function (pdf) of estimation error is relative to a desired pdf. This study firstly proposes extended desirability level metric which has a simpler form compared to an original one. Then a new metric based on principal component analysis is introduced. Illustration examples are given to demonstrate the effectiveness of our proposed measures.

Different estimators have different optimization criteria according to the concrete application considered. Most existing metrics on estimation performance are some averages of estimation errors, which usually give “big” or “small” results to show the “bad” or “good” performance of the evaluated estimators. However, these metrics are only appropriate for measuring minimum mean-square error (MMSE), linear MMSE and even least square estimators and have bias on ones like maximum a posteriori estimators. To handle this problem, a concentration measure is proposed in [1] to measure how concentrative the estimation errors are relative to a desired probability density function. This study proposed several concentration measures including both relative and absolute ones. And the existing concentration measure is extended to more general cases. Illustration examples are provided to verify the effectiveness of our proposed measures.