Development and External Validation of a Gait Test Based Diagnostic Model for Detecting Mild Cognitive Impairment.

OBJECTIVE: To address the lack of large-scale screening tools for mild cognitive impairment (MCI), this study aimed to assess the discriminatory ability of several gait tests for MCI and develop a screening tool based on gait test for MCI. DESIGN: A diagnostic case-control test. SETTING: The general community. PARTICIPANTS: We recruited 134 older adults (≥65 years) for the derivation sample, comprising -69 individuals in the cognitively normal group and -65 in the MCI group (N=134). An additional 70 participants were enrolled for the validation sample. INTERVENTIONS: All participants completed gait tests consisting of a single task (ST) and 3 dual tasks (DTs): counting backwards, serial subtractions 7, and naming animals. MAIN OUTCOME MEASURES: Binary logistic regression analyses were used to develop models, and the efficacy of each model was assessed using receiver operating characteristic (ROC) curve and area under the curve (AUC). The best effective model was the final diagnostic model and validated using ROC curve and calibration curve. RESULTS: The DT gait test incorporating serial subtractions 7 as the cognitive task demonstrated the highest efficacy with the AUC of 0.906 and the accuracy of 0.831 in detecting MCI with "years of education" being adjusted. Furthermore, the model exhibited consistent performance across different age and sex groups. In external validation, the model displayed robust discrimination (AUC=0.913) and calibration (calibrated intercept=-0.062, slope=1.039). CONCLUSIONS: The DT gait test incorporating serial subtractions 7 as the cognitive task demonstrated robust discriminate ability for MCI. This test holds the potential to serve as a large-scale screening tool for MCI, aids in the early detection and intervention of cognitive impairment in older adults.

Predicting Ca2+ and Mg2+ ligand binding sites by deep neural network algorithm.

BACKGROUND: Alkaline earth metal ions are important protein binding ligands in human body, and it is of great significance to predict their binding residues. RESULTS: In this paper, Mg2+ and Ca2+ ligands are taken as the research objects. Based on the characteristic parameters of protein sequences, amino acids, physicochemical characteristics of amino acids and predicted structural information, deep neural network algorithm is used to predict the binding sites of proteins. By optimizing the hyper-parameters of the deep learning algorithm, the prediction results by the fivefold cross-validation are better than those of the Ionseq method. In addition, to further verify the performance of the proposed model, the undersampling data processing method is adopted, and the prediction results on independent test are better than those obtained by the support vector machine algorithm. CONCLUSIONS: An efficient method for predicting Mg2+ and Ca2+ ligand binding sites was presented.

Recognizing five molecular ligand-binding sites with similar chemical structure.

Accurate identification of ligand-binding sites and discovering the protein-ligand interaction mechanism are important for understanding proteins' functions and designing new drugs. Meanwhile, accurate computational prediction and mechanism research are two grand challenges in proteomics. In this article, ligand-binding residues of five ligands (ATP, ADP, GTP, GDP, and NAD) are predicted as a group, due to their similar chemical structures and close biological function relations. The data set of binding sites by five ligands (ATP, ADP, GTP, GDP, and NAD) are collated from Biolip database. Then, five features, containing increment of diversity value, matrix scoring value, auto-covariance, secondary structure information, and surface accessibility information are used in binding site predictions. The support vector machine (SVM) model is used with the five features to predict ligand-binding sites. Finally, prediction results are tested by fivefold cross validation. Accuracy (Acc) of five ligands (ATP, ADP, GTP, GDP, and NAD) achieves 77.4%, 71.2%, 82.1%, 82.9%, and 85.3%, respectively; and Matthew correlation coefficient (MCC) of the above five ligands achieves 0.549, 0.424, 0.643, 0.659, and 0.702, respectively. The research result shows that for ligands with similar chemical structures, microenvironment of their binding sites and their sensitivities to features are similar, while, differences of their ligand-binding properties exist at the same time. © 2019 Wiley Periodicals, Inc.

Recognizing metal and acid radical ion-binding sites by integrating ab initio modeling with template-based transferals.

MOTIVATION: More than half of proteins require binding of metal and acid radical ions for their structure and function. Identification of the ion-binding locations is important for understanding the biological functions of proteins. Due to the small size and high versatility of the metal and acid radical ions, however, computational prediction of their binding sites remains difficult. RESULTS: We proposed a new ligand-specific approach devoted to the binding site prediction of 13 metal ions (Zn2+, Cu2+, Fe2+, Fe3+, Ca2+, Mg2+, Mn2+, Na+, K+) and acid radical ion ligands (CO32-, NO2-, SO42-, PO43-) that are most frequently seen in protein databases. A sequence-based ab initio model is first trained on sequence profiles, where a modified AdaBoost algorithm is extended to balance binding and non-binding residue samples. A composite method IonCom is then developed to combine the ab initio model with multiple threading alignments for further improving the robustness of the binding site predictions. The pipeline was tested using 5-fold cross validations on a comprehensive set of 2,100 non-redundant proteins bound with 3,075 small ion ligands. Significant advantage was demonstrated compared with the state of the art ligand-binding methods including COACH and TargetS for high-accuracy ion-binding site identification. Detailed data analyses show that the major advantage of IonCom lies at the integration of complementary ab initio and template-based components. Ion-specific feature design and binding library selection also contribute to the improvement of small ion ligand binding predictions. AVAILABILITY AND IMPLEMENTATION: http://zhanglab.ccmb.med.umich.edu/IonCom CONTACT: hxz@imut.edu.cn or zhng@umich.eduSupplementary information: Supplementary data are available at Bioinformatics online.

Protein ligand-specific binding residue predictions by an ensemble classifier.

BACKGROUND: Prediction of ligand binding sites is important to elucidate protein functions and is helpful for drug design. Although much progress has been made, many challenges still need to be addressed. Prediction methods need to be carefully developed to account for chemical and structural differences between ligands. RESULTS: In this study, we present ligand-specific methods to predict the binding sites of protein-ligand interactions. First, a sequence-based method is proposed that only extracts features from protein sequence information, including evolutionary conservation scores and predicted structure properties. An improved AdaBoost algorithm is applied to address the serious imbalance problem between the binding and non-binding residues. Then, a combined method is proposed that combines the current template-free method and four other well-established template-based methods. The above two methods predict the ligand binding sites along the sequences using a ligand-specific strategy that contains metal ions, acid radical ions, nucleotides and ferroheme. Testing on a well-established dataset showed that the proposed sequence-based method outperformed the profile-based method by 4-19% in terms of the Matthews correlation coefficient on different ligands. The combined method outperformed each of the individual methods, with an improvement in the average Matthews correlation coefficients of 5.55% over all ligands. The results also show that the ligand-specific methods significantly outperform the general-purpose methods, which confirms the necessity of developing elaborate ligand-specific methods for ligand binding site prediction. CONCLUSIONS: Two efficient ligand-specific binding site predictors are presented. The standalone package is freely available for academic usage at http://dase.ecnu.edu.cn/qwdong/TargetCom/TargetCom_standalone.tar.gz  or request upon the corresponding author.

Association of Life's Simple 7 with mild cognitive impairment in community-dwelling older adults in China: a cross-sectional study.

Background: Life's Simple 7 (LS7), a metric composed of seven intervenable cardiovascular risk factors, is initiated by the American Heart Association to improve cardiovascular health. The components of LS7 have been reported as risk factors for dementia. However, few studies investigated the association between LS7 metric and mild cognitive impairment (MCI). Methods: The study was carried out in a primary care facility between 8 June and 10 July 2022. A total of 297 community-dwelling residents aged 65 years or older were recruited. Sociodemographic, comorbidity, and lifestyle characteristics were collected through the questionnaires, and biological parameters were obtained from blood sample examinations. Logistic regression was used to analyze the association between LS7 scores (overall, behavioral, and biological) and individual components with MCI, adjusting sex, age, education, and cardiovascular disease (CVD). Results: In comparison with the cognitively intact group (n = 195), the MCI group (n = 102) had a lower education level and a higher proportion of hypertension. Multivariate logistic regression analysis, adjusting sex, age, education, and CVD demonstrated a significant association between MCI and overall LS7 score [odd ratio = 0.805, 95% confidence interval (0.690, 0.939)] and biological score [odd ratio = 0.762, 95% confidence interval (0.602, 0.965)]. Conclusion: Life's Simple 7 was associated with MCI in community-dwelling older adults, indicating that LS7 could be used as guidance in the prevention of dementia in the community.

Recognizing protein-metal ion ligands binding residues by random forest algorithm with adding orthogonal properties.

Accurately identifying protein-metal ion ligand binding residues is the key to study protein functions. Because the number of binding residues and non-binding residues is significantly imbalanced, false positives is hard to be eliminated from the binding residues prediction result. Therefore, identification of protein-metal ion ligand binding residues remains challenging. In this paper, the binding site of 7 metal ions (Ca2+, Mg2+, Zn2+, Fe3+, Mn2+, Cu2+ and Co2+) were used as the objects of the study. Besides generally adopted parameters: amino acids and predicted secondary structure information, we creatively introduced ten orthogonal properties as a parameter. These orthogonal properties are clustering of 188 physical and chemical characteristics that can be used to describe three-dimension structural information. With the optimized parameters, we used the Random Forest algorithm to predict ion ligand binding residues. The proposed method obtained good prediction results with the MCC values of Mg2+, Ca2+ and Zn2+ reaching 0.255, 0.254, 0.540, respectively. Comparing to the IonSeq method, the method developed in this paper has advantages on the binding residues prediction of some ions.

Prediction of metal ion ligand binding residues by adding disorder value and propensity factors based on deep learning algorithm.

Proteins need to interact with different ligands to perform their functions. Among the ligands, the metal ion is a major ligand. At present, the prediction of protein metal ion ligand binding residues is a challenge. In this study, we selected Zn2+, Cu2+, Fe2+, Fe3+, Co2+, Mn2+, Ca2+ and Mg2+ metal ion ligands from the BioLip database as the research objects. Based on the amino acids, the physicochemical properties and predicted structural information, we introduced the disorder value as the feature parameter. In addition, based on the component information, position weight matrix and information entropy, we introduced the propensity factor as prediction parameters. Then, we used the deep neural network algorithm for the prediction. Furtherly, we made an optimization for the hyper-parameters of the deep learning algorithm and obtained improved results than the previous IonSeq method.

Prediction of ß-turn types in protein by using composite vector.

Protein secondary structure prediction is an intermediate step in the overall process of tertiary structure prediction. ß-turns are important components of the secondary structure of a protein. Development of an accurate method of prediction of ß-turn types would be helpful for predicting the overall tertiary structure of proteins. In this work, we constructed a database of 2805 protein chains. Our work improved the previous input parameters and used the support vector machine algorithm to predict the ß-turn types; we obtained the overall prediction accuracy of 98.1%, 96.0%, 96.1%, 98.7%, 99.1%, 86.8%, 99.2% and 73.2% with the Matthews Correlation Coefficient values of 0.398, 0.460, 0.043, 0.463, 0.355, 0.172, 0.109 and 0.247, respectively, for types I, II, VIII, I', II', IV, VI and non-ß-turn, respectively. In addition, we also used same method to predict the ß-turn types in three databases of 426, 547 and 823 protein chains and found that our prediction results were better than other predictions.

Recognition of Metal Ion Ligand-Binding Residues by Adding Correlation Features and Propensity Factors.

The realization of many protein functions is inseparable from the interaction with ligands; in particular, the combination of protein and metal ion ligands performs an important biological function. Currently, it is a challenging work to identify the metal ion ligand-binding residues accurately by computational approaches. In this study, we proposed an improved method to predict the binding residues of 10 metal ion ligands (Zn2+, Cu2+, Fe2+, Fe3+, Co2+, Mn2+, Ca2+, Mg2+, Na+, and K+). Based on the basic feature parameters of amino acids, and physicochemical and predicted structural information, we added another two features of amino acid correlation information and binding residue propensity factors. With the optimized parameters, we used the GBM algorithm to predict metal ion ligand-binding residues. In the obtained results, the Sn and MCC values were over 10.17% and 0.297, respectively. Besides, the Sn and MCC values of transition metals were higher than 34.46% and 0.564, respectively. In order to test the validity of our model, another method (Random Forest) was also used in comparison. The better results of this work indicated that the proposed method would be a valuable tool to predict metal ion ligand-binding residues.

Recognition of Ion Ligand Binding Sites Based on Amino Acid Features with the Fusion of Energy, Physicochemical and Structural Features.

[Background: Rational drug molecular design based on virtual screening requires the ligand binding site to be known. Recently, the recognition of ion ligand binding site has become an important research direction in pharmacology. METHODS: In this work, we selected the binding residues of 4 acid radical ion ligands (NO2-, CO32-, SO42- and PO43-) and 10 metal ion ligands (Zn2+, Cu2+, Fe2+, Fe3+, Ca2+, Mg2+, Mn2+, Na+, K+ and Co2+) as research objects. Based on the protein sequence information, we extracted amino acid features, energy, physicochemical, and structure features. Then, we incorporated the above features and input them into the MultilayerPerceptron (MLP) and support vector machine (SVM) algorithms. RESULTS: In the independent test, the best accuracy was higher than 92.5%, which was better than the previous results on the same dataset. In addition, we found that energy information is an important factor affecting the prediction results. CONCLUSION: Finally, we set up a free web server for the prediction of protein-ion ligand binding sites (http://39.104.77.103:8081/lsb/HomePage/HomePage.html). This study is helpful for molecular drug design.

Recognition of beta-hairpin motifs in proteins by using the composite vector.

A composite vector method for predicting beta-hairpin motifs in proteins is proposed by combining the score of matrix, increment of diversity, the value of distance and auto-correlation information to express the information of sequence. The prediction is based on analysis of data from 3,088 non-homologous protein chains including 6,035 beta-hairpin motifs and 2,738 non-beta-hairpin motifs. The overall accuracy of prediction and Matthew's correlation coefficient are 83.1% and 0.59, respectively. In addition, by using the same methods, the accuracy of 80.7% and Matthew's correlation coefficient of 0.61 are obtained for other dataset with 2,878 non-homologous protein chains, which contains 4,884 beta-hairpin motifs and 4,310 non-beta-hairpin motifs. Better results are also obtained in the prediction of the beta-hairpin motifs of proteins by analysis of the CASP6 dataset.

The Identification of Metal Ion Ligand-Binding Residues by Adding the Reclassified Relative Solvent Accessibility.

Many proteins realize their special functions by binding with specific metal ion ligands during a cell's life cycle. The ability to correctly identify metal ion ligand-binding residues is valuable for the human health and the design of molecular drug. Precisely identifying these residues, however, remains challenging work. We have presented an improved computational approach for predicting the binding residues of 10 metal ion ligands (Zn2+, Cu2+, Fe2+, Fe3+, Co2+, Ca2+, Mg2+, Mn2+, Na+, and K+) by adding reclassified relative solvent accessibility (RSA). The best accuracy of fivefold cross-validation was higher than 77.9%, which was about 16% higher than the previous result on the same dataset. It was found that different reclassification of the RSA information can make different contributions to the identification of specific ligand binding residues. Our study has provided an additional understanding of the effect of the RSA on the identification of metal ion ligand binding residues.

Recognizing Ion Ligand-Binding Residues by Random Forest Algorithm Based on Optimized Dihedral Angle.

The prediction of ion ligand-binding residues in protein sequences is a challenging work that contributes to understand the specific functions of proteins in life processes. In this article, we selected binding residues of 14 ion ligands as research objects, including four acid radical ion ligands and 10 metal ion ligands. Based on the amino acid sequence information, we selected the composition and position conservation information of amino acids, the predicted structural information, and physicochemical properties of amino acids as basic feature parameters. We then performed a statistical analysis and reclassification for dihedral angle and proposed new methods on the extraction of feature parameters. The methods mainly included applying information entropy on the extraction of polarization charge and hydrophilic-hydrophobic information of amino acids and using position weight matrices on the extraction of position conservation information. In the prediction model, we used the random forest algorithm and obtained better prediction results than previous works. With the independent test, the Matthew's correlation coefficient and accuracy of 10 metal ion ligand-binding residues were larger than 0.07 and 52%, respectively; the corresponding evaluation values of four acid radical ion ligand-binding residues were larger than 0.15 and 86%, respectively. Further, we classified and combined the phi and psi angles and optimized prediction model for each ion ligand-binding residue.

Prediction of acid radical ion binding residues by K-nearest neighbors classifier.

BACKGROUND: Proteins perform their functions by interacting with acid radical ions. Recently, it was a challenging work to precisely predict the binding residues of acid radical ion ligands in the research field of molecular drug design. RESULTS: In this study, we proposed an improved method to predict the acid radical ion binding residues by using K-nearest Neighbors classifier. Meanwhile, we constructed datasets of four acid radical ion ligand (NO2-, CO32-, SO42-, PO43-) binding residues from BioLip database. Then, based on the optimal window length for each acid radical ion ligand, we refined composition information and position conservative information and extracted them as feature parameters for K-nearest Neighbors classifier. In the results of 5-fold cross-validation, the Matthew's correlation coefficient was higher than 0.45, the values of accuracy, sensitivity and specificity were all higher than 69.2%, and the false positive rate was lower than 30.8%. Further, we also performed an independent test to test the practicability of the proposed method. In the obtained results, the sensitivity was higher than 40.9%, the values of accuracy and specificity were higher than 84.2%, the Matthew's correlation coefficient was higher than 0.116, and the false positive rate was lower than 15.4%. Finally, we identified binding residues of the six metal ion ligands. In the predicted results, the values of accuracy, sensitivity and specificity were all higher than 77.6%, the Matthew's correlation coefficient was higher than 0.6, and the false positive rate was lower than 19.6%. CONCLUSIONS: Taken together, the good results of our prediction method added new insights in the prediction of the binding residues of acid radical ion ligands.

Recognizing ion ligand binding sites by SMO algorithm.

BACKGROUND: In many important life activities, the execution of protein function depends on the interaction between proteins and ligands. As an important protein binding ligand, the identification of the binding site of the ion ligands plays an important role in the study of the protein function. RESULTS: In this study, four acid radical ion ligands (NO2-,CO32-,SO42-,PO43-) and ten metal ion ligands (Zn2+,Cu2+,Fe2+,Fe3+,Ca2+,Mg2+,Mn2+,Na+,K+,Co2+) are selected as the research object, and the Sequential minimal optimization (SMO) algorithm based on sequence information was proposed, better prediction results were obtained by 5-fold cross validation. CONCLUSIONS: An efficient method for predicting ion ligand binding sites was presented.

Electrohydrodynamic Drying of Chinese Wolfberry in a Multiple Needle-to-Plate Electrode System.

In order to systematically and comprehensively investigate electrohydrodynamic (EHD) drying characteristics and mechanisms in a multiple needle-to-plate electrode system, drying experiments of Chinese wolfberry were conducted by blocking ionic wind and changing needle spacing in a multiple needle-to-plate electrode system. Drying characteristics, quality parameters, and the microstructure of Chinese wolfberry fruits were measured. Results show that ionic wind plays a very important role during the drying process. Drying rates of different needle spacing treatments are significantly higher than that of the control, and the drying rate decreases with the increase of needle spacing. Needle spacing has a great influence on the speed of ionic wind, rehydration rate, and polysaccharide contents. The effective moisture diffusion coefficient and the electrical conductivity disintegration index decreases with an increase in needle spacing. Ionic wind has a great influence on the effective moisture diffusion coefficient and the electrical conductivity disintegration index of Chinese wolfberry fruits. The microstructure of Chinese wolfberry fruits dried in an EHD system significantly changed. This study provides a theoretical basis and practical guidance for understanding characteristic parameters and mechanisms of EHD drying technology.

Using support vector machine to predict beta- and gamma-turns in proteins.

By using the composite vector with increment of diversity, position conservation scoring function, and predictive secondary structures to express the information of sequence, a support vector machine (SVM) algorithm for predicting beta- and gamma-turns in the proteins is proposed. The 426 and 320 nonhomologous protein chains described by Guruprasad and Rajkumar (Guruprasad and Rajkumar J. Biosci 2000, 25,143) are used for training and testing the predictive model of the beta- and gamma-turns, respectively. The overall prediction accuracy and the Matthews correlation coefficient in 7-fold cross-validation are 79.8% and 0.47, respectively, for the beta-turns. The overall prediction accuracy in 5-fold cross-validation is 61.0% for the gamma-turns. These results are significantly higher than the other algorithms in the prediction of beta- and gamma-turns using the same datasets. In addition, the 547 and 823 nonhomologous protein chains described by Fuchs and Alix (Fuchs and Alix Proteins: Struct Funct Bioinform 2005, 59, 828) are used for training and testing the predictive model of the beta- and gamma-turns, and better results are obtained. This algorithm may be helpful to improve the performance of protein turns' prediction. To ensure the ability of the SVM method to correctly classify beta-turn and non-beta-turn (gamma-turn and non-gamma-turn), the receiver operating characteristic threshold independent measure curves are provided.

Using feature optimization-based support vector machine method to recognize the ß-hairpin motifs in enzymes.

ß-Hairpins in enzyme, a kind of special protein with catalytic functions, contain many binding sites which are essential for the functions of enzyme. With the increasing number of observed enzyme protein sequences, it is of especial importance to use bioinformatics techniques to quickly and accurately identify the ß-hairpin in enzyme protein for further advanced annotation of structure and function of enzyme. In this work, the proposed method was trained and tested on a non-redundant enzyme ß-hairpin database containing 2818 ß-hairpins and 1098 non-ß-hairpins. With 5-fold cross-validation on the training dataset, the overall accuracy of 90.08% and Matthew's correlation coefficient (Mcc) of 0.74 were obtained, while on the independent test dataset, the overall accuracy of 88.93% and Mcc of 0.76 were achieved. Furthermore, the method was validated on 845 ß-hairpins with ligand binding sites. With 5-fold cross-validation on the training dataset and independent test on the test dataset, the overall accuracies were 85.82% (Mcc of 0.71) and 84.78% (Mcc of 0.70), respectively. With an integration of mRMR feature selection and SVM algorithm, a reasonable high accuracy was achieved, indicating the method to be an effective tool for the further studies of ß-hairpins in enzymes structure. Additionally, as a novelty for function prediction of enzymes, ß-hairpins with ligand binding sites were predicted. Based on this work, a web server was constructed to predict ß-hairpin motifs in enzymes (http://202.207.29.251:8080/).