High-dimensional Genomic Data Research Articles

Machine Learning Techniques and Hybrid Feature Selection Methods for Efficient Prediction of Cancer

The high-dimensional genomic data presents significant challenges, and traditional analytical methods often struggle to capture the complex, non-linear relationships within these datasets. This study elaborates into the application of machine learning methods for dimensionality reduction and predictive modeling of binary phenotypes using gene expression data. Various dimensionality reduction techniques are explored, including t-distributed stochastic neighbor embedding (t-SNE), Non-negative matrix factorization (NMF), Principal component analysis (PCA), and manifold learning methods. Additionally, various algorithms such as logistic regression, random forests, support vector machines (SVMs)cand naive Bayes models are evaluated for predicting phenotypes. The study employs rigorous cross-validation, permutation testing, and evaluation metrics like the Matthews Correlation Coefficient (MCC) to assess model performance. The study rigorously assesses current genomics strategies, pinpointing their drawbacks and suggesting areas for future investigation, while delving into the potential of machine learning to overcome these hurdles and offer valuable insights in genomics. Keywords- Gene Expression , Dimensionality Reduction, Biomarkers, Cancer Prediction, Epigenetics, Machine Learning, Feature extraction.

Multivariate Time Series Change-Point Detection with a Novel Pearson-like Scaled Bregman Divergence.

Change-point detection is a challenging problem that has a number of applications across various real-world domains. The primary objective of CPD is to identify specific time points where the underlying system undergoes transitions between different states, each characterized by its distinct data distribution. Precise identification of change points in time series omics data can provide insights into the dynamic and temporal characteristics inherent to complex biological systems. Many change-point detection methods have traditionally focused on the direct estimation of data distributions. However, these approaches become unrealistic in high-dimensional data analysis. Density ratio methods have emerged as promising approaches for change-point detection since estimating density ratios is easier than directly estimating individual densities. Nevertheless, the divergence measures used in these methods may suffer from numerical instability during computation. Additionally, the most popular -relative Pearson divergence cannot measure the dissimilarity between two distributions of data but a mixture of distributions. To overcome the limitations of existing density ratio-based methods, we propose a novel approach called the Pearson-like scaled-Bregman divergence-based (PLsBD) density ratio estimation method for change-point detection. Our theoretical studies derive an analytical expression for the Pearson-like scaled Bregman divergence using a mixture measure. We integrate the PLsBD with a kernel regression model and apply a random sampling strategy to identify change points in both synthetic data and real-world high-dimensional genomics data of Drosophila. Our PLsBD method demonstrates superior performance compared to many other change-point detection methods.

Application of Deep Learning for the Detection of Genetic Variations: Its Implementation in Classifying Alzheimer's Disease

Deep learning emerges as a promising technique, utilizing nonlinear transformations for feature extraction from high-dimensional datasets. However, its application encounters challenges in genome-wide association studies (GWAS) dealing with high-dimensional genomic data. This study introduces an innovative three-step method termed SWAT-CNN for the identification of genetic variants. This approach employs deep learning to pinpoint phenotype-related single nucleotide polymorphisms (SNPs), facilitating the development of precise disease classification models. In the first step, the entire genome undergoes division into non overlapping fragments of an optimal size. Subsequently, convolutional neural network (CNN) analysis is conducted on each fragment to identify phenotype-associated segments. The second step, employs a Sliding Window Association Test (SWAT), where CNN is utilized on the selected fragments to compute phenotype influence scores (PIS) and detect phenotype-associated SNPs based on these scores. The third step involves running CNN on all identified SNPs to construct a comprehensive classification model. Validation of the proposed approach utilized GWAS data from the Alzheimer’s disease Neuroimaging Initiative (ADNI), encompassing 981 subjects, including cognitively normal older adults (CN) and individuals with Alzheimer's disease (AD). Notably, the method successfully identified the widely recognized APOE region as the most significant genetic locus for AD. The resulting classification model exhibited an area under the curve (AUC) of 0.82, demonstrating compatibility with traditional machine learning approaches such as random forest and XGBoost. SWAT-CNN, as a groundbreaking deep learning-based genome-wide methodology, not only identified AD-associated SNPs but also presented a robust classification model for Alzheimer's disease, suggesting potential applications across diverse biomedical domains.

Modifying the false discovery rate procedure based on the information theory under arbitrary correlation structure and its performance in high-dimensional genomic data

BackgroundControlling the False Discovery Rate (FDR) in Multiple Comparison Procedures (MCPs) has widespread applications in many scientific fields. Previous studies show that the correlation structure between test statistics increases the variance and bias of FDR. The objective of this study is to modify the effect of correlation in MCPs based on the information theory. We proposed three modified procedures (M1, M2, and M3) under strong, moderate, and mild assumptions based on the conditional Fisher Information of the consecutive sorted test statistics for controlling the false discovery rate under arbitrary correlation structure. The performance of the proposed procedures was compared with the Benjamini–Hochberg (BH) and Benjamini–Yekutieli (BY) procedures in simulation study and real high-dimensional data of colorectal cancer gene expressions. In the simulation study, we generated 1000 differential multivariate Gaussian features with different levels of the correlation structure and screened the significance features by the FDR controlling procedures, with strong control on the Family Wise Error Rates.ResultsWhen there was no correlation between 1000 simulated features, the performance of the BH procedure was similar to the three proposed procedures. In low to medium correlation structures the BY procedure is too conservative. The BH procedure is too liberal, and the mean number of screened features was constant at the different levels of the correlation between features. The mean number of screened features by proposed procedures was between BY and BH procedures and reduced when the correlations increased. Where the features are highly correlated the number of screened features by proposed procedures reached the Bonferroni (BF) procedure, as expected. In real data analysis the BY, BH, M1, M2, and M3 procedures were done to screen gene expressions of colorectal cancer. To fit a predictive model based on the screened features the Efficient Bayesian Logistic Regression (EBLR) model was used. The fitted EBLR models based on the screened features by M1 and M2 procedures have minimum entropies and are more efficient than BY and BH procedures.ConclusionThe modified proposed procedures based on information theory, are much more flexible than BH and BY procedures for the amount of correlation between test statistics. The modified procedures avoided screening the non-informative features and so the number of screened features reduced with the increase in the level of correlation.

MVMSGAT: Integrating Multiview, Multi-Scale Graph Convolutional Networks with Biological Prior Knowledge for Predicting Bladder Cancer Response to Neoadjuvant Therapy

The incidence of bladder cancer is on the rise, and its molecular heterogeneity presents significant challenges for personalized cancer therapy. Transcriptome data can characterize the variability among patients. Traditional machine-learning methods often struggle with high-dimensional genomic data, falling into the ’curse of dimensionality’. To address this challenge, we have developed MVMSGAT, an innovative predictive model tailored for forecasting responses to neoadjuvant therapy in bladder cancer patients. MVMSGAT significantly enhances model performance by incorporating multi-perspective biological prior knowledge. It initially utilizes the Boruta algorithm to select key genes from transcriptome data, subsequently constructing a comprehensive graph of gene co-expression and protein–protein interactions. MVMSGAT further employs a graph convolutional neural network to integrate this information within a multiview knowledge graph, amalgamating biological knowledge maps from various scales using an attention mechanism. For validation, MVMSGAT was tested using a five-fold cross-validation approach on two specific GEO datasets, GSE169455 and GSE69795, involving a total of 210 bladder cancer samples. MVMSGAT demonstrated superior performance, with the following metrics (mean ± standard deviation): AUC-ROC of 0.8724±0.0511, accuracy of 0.7789±0.068, F1 score of 0.8529±0.0338, and recall of 0.9231±0.0719. These results underscore the potential of MVMSGAT in advancing personalized treatment and precision medicine in bladder cancer.

Improving predictive ability in sparse testing designs in soybean populations

The availability of high-dimensional genomic data and advancements in genome-based prediction models (GP) have revolutionized and contributed to accelerated genetic gains in soybean breeding programs. GP-based sparse testing is a promising concept that allows increasing the testing capacity of genotypes in environments, of genotypes or environments at a fixed cost, or a substantial reduction of costs at a fixed testing capacity. This study represents the first attempt to implement GP-based sparse testing in soybeans by evaluating different training set compositions going from non-overlapped RILs until almost the other extreme of having same set of genotypes observed across environments for different training set sizes. A total of 1,755 recombinant inbred lines (RILs) tested in nine environments were used in this study. RILs were derived from 39 bi-parental populations of the Soybean Nested Association Mapping (NAM) project. The predictive abilities of various models and training set sizes and compositions were investigated. Training compositions included a range of ratios of overlapping (O-RILs) and non-overlapping (NO-RILs) RILs across environments, as well as a methodology to maximize or minimize the genetic diversity in a fixed-size sample. Reducing the training set size compromised predictive ability in most training set compositions. Overall, maximizing the genetic diversity within the training set and the inclusion of O-RILs increased prediction accuracy given a fixed training set size; however, the most complex model was less affected by these factors. More testing environments in the early stages of the breeding pipeline can provide a more comprehensive assessment of genotype stability and adaptation which are fundamental for the precise selection of superior genotypes adapted to a wide range of environments.

Characterizing the Survival-Associated Interactions between Tumor-infiltrating Lymphocytes and Tumors from Pathological Images and Multi-omics Data.

The tumor-infiltrating lymphocytes (TILs) and its correlation with tumors have shown significant values in the development of cancers. Many observations indicated that the combination of the whole-slide pathological images (WSIs) and genomic data can better characterize the immunological mechanisms of TILs. However, the existing image-genomic studies evaluated the TILs by the combination of pathological image and single-type of omics data (e.g., mRNA), which is difficulty in assessing the underlying molecular processes of TILs holistically. Additionally, it is still very challenging to characterize the intersections between TILs and tumor regions in WSIs and the high dimensional genomic data also brings difficulty for the integrative analysis with WSIs. Based on the above considerations, we proposed an end-to-end deep learning framework i.e., IMO-TILs that can integrate pathological image with multi-omics data (i.e., mRNA and miRNA) to analyze TILs and explore the survival-associated interactions between TILs and tumors. Specifically, we firstly apply the graph attention network to describe the spatial interactions between TILs and tumor regions in WSIs. As to genomic data, the Concrete AutoEncoder (i.e., CAE) is adopted to select survival-associated Eigengenes from the high-dimensional multi-omics data. Finally, the deep generalized canonical correlation analysis (DGCCA) accompanied with the attention layer is implemented to fuse the image and multi-omics data for prognosis prediction of human cancers. The experimental results on three cancer cohorts derived from the Cancer Genome Atlas (TCGA) indicated that our method can both achieve higher prognosis results and identify consistent imaging and multi-omics bio-markers correlated strongly with the prognosis of human cancers.

Artificial intelligence-driven biomedical genomics

As genomic research becomes more complex and data-rich, artificial intelligence (AI) has emerged as a crucial tool for processing and analyzing high-dimensional genomic data, accelerating biomarker discovery, and enhancing genomic sequence annotations. Despite the increasing application of AI in genomic research, challenges persist, particularly regarding the integration of biomedical knowledge into algorithm development. We reviewed high-quality, AI-driven biomedical genomic studies from the past five years, covering applications in disease prediction, detection, diagnosis, and treatment. Each category highlights how different AI techniques are applied in biomedical contexts. Furthermore, we identify current challenges and potential solutions in AI-assisted biomedical genomics. This comprehensive review is designed to encourage collaboration among computer scientists, healthcare professionals, and interested communities, propelling the development of AI applications that can be smoothly integrated into routine medical services.

Utilizing Graph Neural Networks for Breast Cancer Prognosis Prediction with High-dimensional Genomic Data.

An accurate prediction of breast cancer is essential to help physicians make appropriate treatment recommendations to reduce the chance of excessive treatment, avoiding unnecessary anxiety for patients. Cancer prognosis is highly related to patients' genomic features, which are high-dimensional in nature. In this study, we utilize a systems biology feature selector for dimension reduction to select 20 prognostic biomarkers that are considered closely related to breast cancer prognosis from the high dimensional RNA Sequencing (RNA-Seq) data. Furthermore, we establish a graph neural network (GNN) and a multi-layer perception (MLP) graph-level readout method to better extract the underlying gene interactions from the corresponding gene interaction network (GIN). With the help of GINs, the model performs the best among all baseline models, especially in the area under the precision-recall curve (AUPRC) by as large as 23%. The results demonstrate that our approach using GNNs can successfully extract high-dimensional and complicated interactions within genomic data.

Bayesian weighted random forest for classification of high-dimensional genomics data

In this paper, a full Bayesian weighted probabilistic model is developed for random classification trees. The new model Bayesian Weighted Random Classification Forest (BWRCF) arises from the modification of the existing random classification forest in two ways. Firstly, the tree terminal node estimation procedure is replaced with a Bayesian estimation approach. Secondly, a new variable ranking procedure is developed and then hybridized with BWRCF to tackle the high-dimensionality issues. The performance of the proposed method is analyzed using simulated and real-life high-dimensional microarray datasets based on holdout accuracy and misclassification error rates. The results of the analyses showed that the proposed BWRCF is robust in terms of its ability to withstand moderate to large high-dimensionality scenarios. In addition, BWRCF also has improved predictive and efficiency abilities over selected competing methods.

ForestSubtype: a cancer subtype identifying approach based on high-dimensional genomic data and a parallel random forest

BackgroundCancer subtype classification is helpful for personalized cancer treatment. Although, some approaches have been developed to classifying caner subtype based on high dimensional gene expression data, it is difficult to obtain satisfactory classification results. Meanwhile, some cancers have been well studied and classified to some subtypes, which are adopt by most researchers. Hence, this priori knowledge is significant for further identifying new meaningful subtypes.ResultsIn this paper, we present a combined parallel random forest and autoencoder approach for cancer subtype identification based on high dimensional gene expression data, ForestSubtype. ForestSubtype first adopts the parallel RF and the priori knowledge of cancer subtype to train a module and extract significant candidate features. Second, ForestSubtype uses a random forest as the base module and ten parallel random forests to compute each feature weight and rank them separately. Then, the intersection of the features with the larger weights output by the ten parallel random forests is taken as our subsequent candidate features. Third, ForestSubtype uses an autoencoder to condenses the selected features into a two-dimensional data. Fourth, ForestSubtype utilizes k-means++ to obtain new cancer subtype identification results. In this paper, the breast cancer gene expression data obtained from The Cancer Genome Atlas are used for training and validation, and an independent breast cancer dataset from the Molecular Taxonomy of Breast Cancer International Consortium is used for testing. Additionally, we use two other cancer datasets for validating the generalizability of ForestSubtype. ForestSubtype outperforms the other two methods in terms of the distribution of clusters, internal and external metric results. The open-source code is available at https://github.com/lffyd/ForestSubtype.ConclusionsOur work shows that the combination of high-dimensional gene expression data and parallel random forests and autoencoder, guided by a priori knowledge, can identify new subtypes more effectively than existing methods of cancer subtype classification.

Multiset correlation and factor analysis enables exploration of multi-omics data

Multi-omics datasets are becoming more common, necessitating better integration methods to realize their revolutionary potential. Here, we introduce multi-set correlation and factor analysis (MCFA), an unsupervised integration method tailored to the unique challenges of high-dimensional genomics data that enables fast inference of shared and private factors. We used MCFA to integrate methylation markers, protein expression, RNA expression, and metabolite levels in 614 diverse samples from the Trans-Omics for Precision Medicine/Multi-Ethnic Study of Atherosclerosis multi-omics pilot. Samples cluster strongly by ancestry in the shared space, even in the absence of genetic information, while private spaces frequently capture dataset-specific technical variation. Finally, we integrated genetic data by conducting a genome-wide association study (GWAS) of our inferred factors, observing that several factors are enriched for GWAS hits and trans-expression quantitative trait loci. Two of these factors appear to be related to metabolic disease. Our study provides a foundation and framework for further integrative analysis of ever larger multi-modal genomic datasets.

Stable Variable Selection for High-Dimensional Genomic Data with Strong Correlations

Stable Variable Selection for High-Dimensional Genomic Data with Strong Correlations

Deep Learning Techniques with Genomic Data in Cancer Prognosis: A Comprehensive Review of the 2021-2023 Literature.

Deep learning has brought about a significant transformation in machine learning, leading to an array of novel methodologies and consequently broadening its influence. The application of deep learning in various sectors, especially biomedical data analysis, has initiated a period filled with noteworthy scientific developments. This trend has majorly influenced cancer prognosis, where the interpretation of genomic data for survival analysis has become a central research focus. The capacity of deep learning to decode intricate patterns embedded within high-dimensional genomic data has provoked a paradigm shift in our understanding of cancer survival. Given the swift progression in this field, there is an urgent need for a comprehensive review that focuses on the most influential studies from 2021 to 2023. This review, through its careful selection and thorough exploration of dominant trends and methodologies, strives to fulfill this need. The paper aims to enhance our existing understanding of applications of deep learning in cancer survival analysis, while also highlighting promising directions for future research. This paper undertakes aims to enrich our existing grasp of the application of deep learning in cancer survival analysis, while concurrently shedding light on promising directions for future research in this vibrant and rapidly proliferating field.

Geometric graph neural networks on multi-omics data to predict cancer survival outcomes

The advance of sequencing technologies has enabled a thorough molecular characterization of the genome in human cancers. To improve patient prognosis predictions and subsequent treatment strategies, it is imperative to develop advanced computational methods to analyze large-scale, high-dimensional genomic data. However, traditional machine learning methods face a challenge in handling the high-dimensional, low-sample size problem that is shown in most genomic data sets. To address this, our group has developed geometric network analysis techniques on multi-omics data in connection with prior biological knowledge derived from protein-protein interactions (PPIs) or pathways. Geometric features obtained from the genomic network, such as Ollivier-Ricci curvature and the invariant measure of the associated Markov chain, have been shown to be predictive of survival outcomes in various cancers. In this study, we propose a novel supervised deep learning method called geometric graph neural network (GGNN) that incorporates such geometric features into deep learning for enhanced predictive power and interpretability. More specifically, we utilize a state-of-the-art graph neural network with sparse connections between the hidden layers based on known biology of the PPI network and pathway information. Geometric features along with multi-omics data are then incorporated into the corresponding layers. The proposed approach utilizes a local–global principle in such a manner that highly predictive features are selected at the front layers and fed directly to the last layer for multivariable Cox proportional-hazards regression modeling. The method was applied to multi-omics data from the CoMMpass study of multiple myeloma and ten major cancers in The Cancer Genome Atlas (TCGA). In most experiments, our method showed superior predictive performance compared to other alternative methods.

Overlapping group screening for binary cancer classification with TCGA high-dimensional genomic data.

Precision medicine has been a global trend of medical development, wherein cancer diagnosis plays an important role. With accurate diagnosis of cancer, we can provide patients with appropriate medical treatments for improving patients' survival. Since disease developments involve complex interplay among multiple factors such as gene-gene interactions, cancer classifications based on microarray gene expression profiling data are expected to be effective, and hence, have attracted extensive attention in computational biology and medicine. However, when using genomic data to build a diagnostic model, there exist several problems to be overcome, including the high-dimensional feature space and feature contamination. In this paper, we propose using the overlapping group screening (OGS) approach to build an accurate cancer diagnosis model and predict the probability of a patient falling into some disease classification category in the logistic regression framework. This new proposal integrates gene pathway information into the procedure for identifying genes and gene-gene interactions associated with the classification of cancer outcome groups. We conduct a series of simulation studies to compare the predictive accuracy of our proposed method for cancer diagnosis with some existing machine learning methods, and find the better performances of the former method. We apply the proposed method to the genomic data of The Cancer Genome Atlas related to lung adenocarcinoma (LUAD), liver hepatocellular carcinoma (LHC), and thyroid carcinoma (THCA), to establish accurate cancer diagnosis models.

Genomics data analysis via spectral shape and topology.

Mapper, a topological algorithm, is frequently used as an exploratory tool to build a graphical representation of data. This representation can help to gain a better understanding of the intrinsic shape of high-dimensional genomic data and to retain information that may be lost using standard dimension-reduction algorithms. We propose a novel workflow to process and analyze RNA-seq data from tumor and healthy subjects integrating Mapper, differential gene expression, and spectral shape analysis. Precisely, we show that a Gaussian mixture approximation method can be used to produce graphical structures that successfully separate tumor and healthy subjects, and produce two subgroups of tumor subjects. A further analysis using DESeq2, a popular tool for the detection of differentially expressed genes, shows that these two subgroups of tumor cells bear two distinct gene regulations, suggesting two discrete paths for forming lung cancer, which could not be highlighted by other popular clustering methods, including t-distributed stochastic neighbor embedding (t-SNE). Although Mapper shows promise in analyzing high-dimensional data, tools to statistically analyze Mapper graphical structures are limited in the existing literature. In this paper, we develop a scoring method using heat kernel signatures that provides an empirical setting for statistical inferences such as hypothesis testing, sensitivity analysis, and correlation analysis.

Deep learning-based polygenic risk analysis for Alzheimer’s disease prediction

BackgroundThe polygenic nature of Alzheimer’s disease (AD) suggests that multiple variants jointly contribute to disease susceptibility. As an individual’s genetic variants are constant throughout life, evaluating the combined effects of multiple disease-associated genetic risks enables reliable AD risk prediction. Because of the complexity of genomic data, current statistical analyses cannot comprehensively capture the polygenic risk of AD, resulting in unsatisfactory disease risk prediction. However, deep learning methods, which capture nonlinearity within high-dimensional genomic data, may enable more accurate disease risk prediction and improve our understanding of AD etiology. Accordingly, we developed deep learning neural network models for modeling AD polygenic risk.MethodsWe constructed neural network models to model AD polygenic risk and compared them with the widely used weighted polygenic risk score and lasso models. We conducted robust linear regression analysis to investigate the relationship between the AD polygenic risk derived from deep learning methods and AD endophenotypes (i.e., plasma biomarkers and individual cognitive performance). We stratified individuals by applying unsupervised clustering to the outputs from the hidden layers of the neural network model.ResultsThe deep learning models outperform other statistical models for modeling AD risk. Moreover, the polygenic risk derived from the deep learning models enables the identification of disease-associated biological pathways and the stratification of individuals according to distinct pathological mechanisms.ConclusionOur results suggest that deep learning methods are effective for modeling the genetic risks of AD and other diseases, classifying disease risks, and uncovering disease mechanisms.

Boosting multivariate structured additive distributional regression models.

We develop a model-based boosting approach for multivariate distributional regression within the framework of generalized additive models for location, scale, and shape. Our approach enables the simultaneous modeling of all distribution parameters of an arbitrary parametric distribution of a multivariate response conditional on explanatory variables, while being applicable to potentially high-dimensional data. Moreover, the boosting algorithm incorporates data-driven variable selection, taking various different types of effects into account. As a special merit of our approach, it allows for modeling the association between multiple continuous or discrete outcomes through the relevant covariates. After a detailed simulation study investigating estimation and prediction performance, we demonstrate the full flexibility of our approach in three diverse biomedical applications. The first is based on high-dimensional genomic cohort data from the UK Biobank, considering a bivariate binary response (chronic ischemic heart disease and high cholesterol). Here, we are able to identify genetic variants that are informative for the association between cholesterol and heart disease. The second application considers the demand for health care in Australia with the number of consultations and the number of prescribed medications as a bivariate count response. The third application analyses two dimensions of childhood undernutrition in Nigeria as a bivariate response and we find that the correlation between the two undernutrition scores is considerably different depending on the child's age and the region the child livesin.

Improving the Post-Operative Prediction of BCR-Free Survival Time with mRNA Variables and Machine Learning

Predicting the risk of, and time to biochemical recurrence (BCR) in prostate cancer patients post-operatively is critical in patient treatment decision pathways following surgical intervention. This study aimed to investigate the predictive potential of mRNA information to improve upon reference nomograms and clinical-only models, using a dataset of 187 patients that includes over 20,000 features. Several machine learning methodologies were implemented for the analysis of censored patient follow-up information with such high-dimensional genomic data. Our findings demonstrated the potential of inclusion of mRNA information for BCR-free survival prediction. A random survival forest pipeline was found to achieve high predictive performance with respect to discrimination, calibration, and net benefit. Two mRNA variables, namely ESM1 and DHAH8, were identified as consistently strong predictors with this dataset.