An inferential framework for biological network hypothesis tests

An established way for improving the accuracy of gridded satellite precipitation products is to “correct” them by exploiting ground-based precipitation measurements, together with machine and statistical learning algorithms. Such corrections are made in regression settings, where the ground-based measurements are the dependent variable and the satellite data are predictor variables. Comparisons of machine and statistical learning algorithms in the direction of obtaining the most useful precipitation datasets by performing such corrections are regularly conducted in the literature. Nonetheless, in most of these comparisons, a small number of machine and statistical learning algorithms are considered. Also, small geographical regions and limited time periods are examined. Thus, the results provided tend to be of local importance and to not offer more general guidance. To provide results that are generalizable, we compared eight state-of-the-art machine and statistical learning algorithms in correcting satellite precipitation data for the entire contiguous United States and for a 15-year period. We used monthly data from the PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) gridded dataset and the Global Historical Climatology Network monthly database, version 2 (GHCNm). Our results suggest that extreme gradient boosting (XGBoost) and random forests are more accurate than the remaining algorithms, which can be ordered as follows from the best to the worst ones: Bayesian regularized feed-forward neural networks, multivariate adaptive polynomial splines (poly-MARS), gradient boosting machines (gbm), multivariate adaptive regression splines (MARS), feed-forward neural networks, linear regression.

BackgroundDespite the interest in machine learning (ML) algorithms for analyzing real-world data (RWD) in healthcare, the use of ML in predicting time-to-event data, a common scenario in clinical practice, is less explored. ML models are capable of algorithmically learning from large, complex datasets and can offer advantages in predicting time-to-event data. We reviewed the recent applications of ML for survival analysis using RWD in healthcare.MethodsPUBMED and EMBASE were searched from database inception through March 2023 to identify peer-reviewed English-language studies of ML models for predicting time-to-event outcomes using the RWD. Two reviewers extracted information on the data source, patient population, survival outcome, ML algorithms, and the Area Under the Curve (AUC).ResultsOf 257 citations, 28 publications were included. Random survival forests (N = 16, 57%) and neural networks (N = 11, 39%) were the most popular ML algorithms. There was variability across AUC for these ML models (median 0.789, range 0.6–0.950). ML algorithms were predominately considered for predicting overall survival in oncology (N = 12, 43%). ML survival models were often used to predict disease prognosis or clinical events (N = 27, 96%) in the oncology, while less were used for treatment outcomes (N = 1, 4%).ConclusionsThe ML algorithms, random survival forests and neural networks, are mainly used for RWD to predict survival outcomes such as disease prognosis or clinical events in the oncology. This review shows that more opportunities remain to apply these ML algorithms to inform treatment decision-making in clinical practice. More methodological work is also needed to ensure the utility and applicability of ML models in survival outcomes.

Understanding transit ridership in an equity context through a comparison of statistical and machine learning algorithms

The high likelihood of noise and the lack of labels in Internet of Things (IoT) time-series sensor readings make the detection of abnormalities an important subject of study. When there is a high correlation between sensor data points, conventional data mining and machine learning algorithms are unable to detect abnormalities. Furthermore, the amount and speed with which IoT sensors report data is one of the reasons why traditional machine and statistical learning algorithms are unable to detect anomalies. In recent years, Artificial Neural Networks (ANN) have been widely used for developing Deep Learning models for learning and classifying unlabeled data for detecting anomalies with high accuracy. In this connection, this paper presents a study on the performance of two well-known ANN models, such as Generative Adversarial Network (GAN) and Variational Auto Encoder (VAE), and a classification model, One Class Support Vector Machine (OCSVM). To evaluate performance, simulations were conducted using a well-known network and sensor datasets. The outcomes show which technique outperformed the others in terms of detection accuracy and training duration.

Network models with latent geometry have been used successfully in many applications in network science and other disciplines, yet it is usually impossible to tell if a given real network is geometric, meaning if it is a typical element in an ensemble of random geometric graphs. Here we identify structural properties of networks that guarantee that random graphs having these properties are geometric. Specifically we show that random graphs in which expected degree and clustering of every node are fixed to some constants are equivalent to random geometric graphs on the real line, if clustering is sufficiently strong. Large numbers of triangles, homogeneously distributed across all nodes as in real networks, are thus a consequence of network geometricity. The methods we use to prove this are quite general and applicable to other network ensembles, geometric or not, and to certain problems in quantum gravity.

In response to rising rates of stress-related psychopathology, prevention programs have proliferated. These programs often target young adults, as the transition to adulthood marks a sensitive period for the emergence of psychiatric disorders. These programs typically target multiple processes and are therefore time-consuming. Here, we seek to advance our understanding of the relationships between multiple target processes and well-being as a first step toward simplifying these trainings to facilitate dissemination and scalability. We assessed common prevention targets (i.e., emotion regulation strategy use, reward responsiveness, working memory, and delay discounting) and subjective well-being in 463 healthy young adults. First, we utilized a regularized partial correlation network to estimate relationships between target processes and subjective well-being. Subsequently, we applied a directed acyclic graph (DAG) with a Bayesian hill-climbing algorithm to infer potential causal relationships between these constructs. Working memory was a key node associated with well-being across network models. Other nodes associated with well-being in the partial correlation network (i.e., reward responsiveness, reappraisal) appeared to be downstream consequences of well-being in the DAG. Though well-being was associated with greater habitual use of reappraisal and lower habitual use of rumination and catastrophizing in the partial correlation network, emotion regulation strategies were not directly linked to well-being in the DAG. Using a Bayesian network approach, our findings point toward working memory as a potential key mechanism of well-being. Future work is needed to replicate these findings and determine the robustness of the identified network.

Microarray data often contains multiple missing genetic expression values that degrade the performance of statistical and machine learning algorithms. This paper presents a K ranked diagonal covariance-based missing value estimation algorithm (KRCOV) that has demonstrated significantly superior performance compared to the more commonly used K-nearest neighbour (KNN) imputation algorithm when it is applied to estimate missing values of BRCA1, BRCA2 and Sporadic genetic mutation samples present in ovarian cancer. Experimental results confirm KRCOV outperformed both KNN and zero imputation techniques in terms of their classification accuracies when used toimpute randomly missing values from 1% to 5%.The classifier used for this purpose was the Generalized Regression Neural Network.The paper also provides a hypothesis for why KRCOV performs better than KNN not only for bioinformatics data but also for other data types having strong correlated values.

This paper first constructs a tiered network model of the interbank market. Then, from the perspective of contagion risk, it studies numerically the resilience of four types of interbank market network models to shocks, namely, tiered networks, random networks, small‐world networks, and scale‐free networks. This paper studies the interbank market with homogeneous and heterogeneous banks and analyzes random shocks and selective shocks. The study reveals that tiered interbank market networks and random interbank market networks are basically more vulnerable against selective shocks, while small‐world interbank market networks and scale‐free interbank market networks are generally more vulnerable against random shocks. Besides, the results indicate that, in the four types of interbank market networks, scale‐free networks have the highest stability against shocks, while small‐world networks are the most vulnerable. When banks are homogeneous, faced with selective shocks, the stability of the tiered interbank market networks is slightly lower than that of random interbank market networks, whereas, in other cases, the stability of the tiered interbank market networks is basically between that of random interbank market networks and that of scale‐free interbank market networks.

Analysis of high-throughput proteomic/genomic data, in particular, surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) data and microarray data, has led to a multitude of techniques aimed at identifying potential biomarkers. Most of the statistical techniques for comparing two groups are based on qualitative measures such as P-value. A quantitative way such as interval estimation for the contrasts of two groups is more appealing. We have devised a simultaneous confidence bands method capable of detecting potential biomarkers, while controlling for overall confidence coverage level, in high-dimensional datasets that discriminate two treatment groups using a permutation scheme. For example, for the SELDI-TOF MS data, we deal with the entire spectrum simultaneously and construct (1 - alpha) confidence bands for the mean differences between groups. Furthermore, peaks were identified based on the maximal differences between the groups as determined by the confidence bands. The analysis method herein described gives both qualitative (P-value) and quantitative data (magnitude of difference). The Clinical Proteomics Programs Databank's ovarian cancer dataset and data from in-house samples containing known spiked-in proteins were analyzed. We were able to identify potential biomarkers similar to those described in previous analysis of the ovarian cancer data, however, while these markers are highly significant between cancer and normal groups, our analysis indicated the absolute difference between the two groups was minimal. In addition, we found additional markers than those previously described with greater differences in average intensities. The proposed confidence bands method successfully detected the spiked-in peaks, as well as, secondary peaks generated by adducts and double-charged species. We also illustrate our method utilizing paired gene expression data from a prostate cancer microarray experiment by constructing confidence bands for the fold changes between cancer and normal samples. R-package, 'seie.zip' (license: GNU GPL), is publiclly available at http://research2.dfci.harvard.edu/dfci/MS_spike-in_data/

The evolution of machine learning (ML) algorithms provides researchers and engineers with state-of-the-art tools to dynamically model complex relationships. The design and operation of municipal solid waste (MSW) management systems require accurate estimation of generation rates. In this study, we applied rapid, non-linear and non-parametric data driven ML algorithms independently, multi-layer perceptron artificial neural network (MLP-ANN) and support vector regression (SVR) models to predict annual MSW generation rates in Bahrain. Models were trained and tested with MSW generation data for period of 1997-2019. The population, gross domestic product, annual tourist numbers, annual electricity consumption and total annual CO2 emissions were selected as explanatory variables and incorporated into developed models. The zero score normalization (ZSN) and minimum maximum normalization (MMN) methods were utilized to improve the quality of data and subsequently enhances the performance of ML algorithms. Statistical metrics were employed to discriminate performance of MLP-ANN and SVR models. The linear, polynomial, radial basis function (RBF) and sigmoid kernel functions were investigated to find the optimal SVR model. Results showed that RBF-SVR model with R2 value of 0.97% and 4.82% and absolute forecasting error (AFE) for the period of 2008 and 2019 exhibits superior prediction and robustness in comparison to MLP-ANN. The efficacy of MLP-ANN model was also reasonably successful with R2 value of 0.94. It was shown that MMN pre-processing generated optimal MLP-ANN model while ZSN pre-processing produced optimal RBF-SVR model. This work also highlights the importance of application of ML modelling approaches to plan and implement their roadmap for waste management systems by policymakers.

In this chapter, we give basic definitions related to random variable networks. After rigorous definitions of a random variable network and a network model, we give definitions of specific network structures: maximum spanning tree, planar maximally filtered graph, concentration graph, threshold graph, maximum clique, and maximum independent set in the threshold graph. All definitions are illustrated by examples. Then we consider a large class of network models generated by different distributions and different measures of similarity. For distributions, we use an important class of elliptical distributions. For pairwise measures of similarity, we consider Pearson correlation, Kruskal correlation, sign similarity, Fechner correlation, Kendall and Spearman correlations. As a main result of the chapter, we establish connections between network structures in different random variable network models. In addition, we define the partial correlation network and establish its properties for elliptical distributions.

Addictive-type sexual behaviour results in significant distress and impact to daily functioning. Furthermore, differences have emerged regarding behavioural presentation and psychosocial outcomes between heterosexual men, heterosexual women, and gender and sexual minority (GSM) individuals. The current research employed partial correlation and Bayesian network modelling, analysing differences in symptoms of distress and sexual addiction among 934 adults aged 18-64 years (529 heterosexual men, M ­ age = 29.9; 211 heterosexual women, M age = 31.8; 194 GSM individuals, M age = 26.7). Sex addiction was measured using the Bergen-Yale Sex Addiction Scale (BYSAS) and distress operationalised by the Depression, Anxiety, Stress Scale (DASS-21). Distinct differences emerged. Heterosexual men may use sex to cope with underlying symptoms of depression, while GSM individuals may use sex as a means of coping with stress and anxiety. Heterosexual women displayed lower scores for sexual addiction and were more likely to endorse attempts at reducing sexual behaviour prior to experiencing consequences. Within-group analyses between non-heterosexual men and women largely converged with network models, revealing elevated symptoms of sexual addiction and distress relative to cisgender heterosexual individuals. The current study highlights the importance of intersectionality in the treatment, assessment, and continued research related to addictive-type sexual behaviour.

Student outcomes are of great importance in higher education institutions. Accreditation bodies focus on them as an indicator to measure the performance and effectiveness of the institution. Forecasting students’ academic performance is crucial for every educational establishment seeking to enhance performance and perseverance of its students and reduce the failure rate in the future. The main goal of this study is to predict the performance of undergraduate first-level students in the Computer Department during the years 2016 to 2021 to enhance their performance in future by discovering the best algorithm use to analyze the educational data to identify the students’ academic performance. The secondary data was collected by reviewing the Student Affairs Department at the Faculty of Specific Education at Damietta University, in addition to the Statistics Department at the university. The dataset contained 830 instances after excluding 139 instances of missing values, irrelevant rows, and outliers. The dataset was divided into train (577 instances (70%)), test (253 instances (30%)) and involved six features such year, midterm, practical exam, writing exam, final total degree, and grade. This paper use five machine learning (ML) algorithms which was selected according to the literature review and high accuracy in predicting educational data mining: For the purpose of comparison, a number of different machine learning algorithms, such as Random Forest, Decision Tree, Naive Bayes, Neural Network, and K-Nearest Neighbours, were utilized and evaluated with evaluation metrics such as confusion matrix, accuracy, precision, recall, and F-measure. The Random Forest and Decision Tree classifiers emerged as the top-performing algorithms, accurately categorizing 250 instances when predicting students' performance in the statistics course. This was determined based on the findings of the study. Out of a total of 253 instances that were included in the testing set, they only made three incorrect classifications.

ABSTRACTThe interrelationships between the symptom domains of dissociation, such as the loss of continuity in subjective experience, the inability to access personal information, and the distortions about the perception of self and the environment, need to be better understood. In the current study, 2274 adults from Italy completed the Dissociative Experiences Scale-II (DES-II), and their responses were examined within a correlation network analysis framework. Fifteen dissociative experiences showed the strongest associations with the other dissociative experiences included in the measure, and they were selected for further analysis. A partial correlation network was calculated to reveal the associations between such experiences, and a community detection analysis was used to explore whether they formed distinct clusters in the network. Subsequently, a Bayesian network was estimated to examine the direction of the associations among the dissociative experiences, and a directed acyclic graph (DAG) was generated to estimate a potentially causal model of their relationships. The community detection analysis revealed three clusters of experiences that were conceptualized in terms of trance, experiential disconnectedness, and segregated behaviors. Dissociative amnesia was a common denominator of all the three clusters. The analysis of the DAG further suggested that dissociation can be conceptualized as a network in which dissociative experiences are layered into groups of symptoms that interact among them. Cognizance of the configuration and interactions among the dissociative domains and their related symptoms may be critical for better understanding the internal logic behind the dissociative processes and for addressing them effectively in clinical practice.

Mild Cognitive Impairment (MCI) is a condition characterized by a decline in cognitive abilities, specifically in memory, language, and attention, that is beyond what is expected due to normal aging. Detection of MCI is crucial for providing appropriate interventions and slowing down the progression of dementia. There are several automated predictive algorithms for prediction using time-to-event data, but it is not clear which is best to predict the time to conversion to MCI. There is also confusion if algorithms with fewer training weights are less accurate. We compared three algorithms, from smaller to large numbers of training weights: a statistical predictive model (Cox proportional hazards model, CoxPH), a machine learning model (Random Survival Forest, RSF), and a deep learning model (DeepSurv). To compare the algorithms under different scenarios, we created a simulated dataset based on the Alzheimer NACC dataset. We found that the CoxPH model was among the best-performing models, in all simulated scenarios. In a larger sample size (n = 6,000), the deep learning algorithm (DeepSurv) exhibited comparable accuracy (73.1%) to the CoxPH model (73%). In the past, ignoring heterogeneity in the CoxPH model led to the conclusion that deep learning methods are superior. We found that when using the CoxPH model with heterogeneity, its accuracy is comparable to that of DeepSurv and RSF. Furthermore, when unobserved heterogeneity is present, such as missing features in the training, all three models showed a similar drop in accuracy. This simulation study suggests that in some applications an algorithm with a smaller number of training weights is not disadvantaged in terms of accuracy. Since algorithms with fewer weights are inherently easier to explain, this study can help artificial intelligence research develop a principled approach to comparing statistical, machine learning, and deep learning algorithms for time-to-event predictions.