A kernel machine method for detecting higher order interactions in multimodal datasets: Application to schizophrenia

TreeKernel: interpretable kernel machine tests for interactions between -omics and clinical predictors with applications to metabolomics and COPD phenotypes

BACKGROUND

In this paper, we are interested in interactions between a high-dimensional -omics dataset and clinical covariates. The goal is to evaluate the relationship between a phenotype of interest and a high-dimensional omics pathway, where the effect of the omics data depends on subjects' clinical covariates (age, sex, smoking status, etc.). For instance, metabolic pathways can vary greatly between sexes which may also change the relationship between certain metabolic pathways and a clinical phenotype of interest. We propose partitioning the clinical covariate space and performing a kernel association test within those partitions. To illustrate this idea, we focus on hierarchical partitions of the clinical covariate space and kernel tests on metabolic pathways.

RESULTS

We see that our proposed method outperforms competing methods in most simulation scenarios. It can identify different relationships among clinical groups with higher power in most scenarios while maintaining a proper Type I error rate. The simulation studies also show a robustness to the grouping structure within the clinical space. We also apply the method to the COPDGene study and find several clinically meaningful interactions between metabolic pathways, the clinical space, and lung function.

CONCLUSION

TreeKernel provides a simple and interpretable process for testing for relationships between high-dimensional omics data and clinical outcomes in the presence of interactions within clinical cohorts. The method is broadly applicable to many studies.

Multimodal machine learning in precision health: A scoping review

Machine learning is frequently being leveraged to tackle problems in the health sector including utilization for clinical decision-support. Its use has historically been focused on single modal data. Attempts to improve prediction and mimic the multimodal nature of clinical expert decision-making has been met in the biomedical field of machine learning by fusing disparate data. This review was conducted to summarize the current studies in this field and identify topics ripe for future research. We conducted this review in accordance with the PRISMA extension for Scoping Reviews to characterize multi-modal data fusion in health. Search strings were established and used in databases: PubMed, Google Scholar, and IEEEXplore from 2011 to 2021. A final set of 128 articles were included in the analysis. The most common health areas utilizing multi-modal methods were neurology and oncology. Early fusion was the most common data merging strategy. Notably, there was an improvement in predictive performance when using data fusion. Lacking from the papers were clear clinical deployment strategies, FDA-approval, and analysis of how using multimodal approaches from diverse sub-populations may improve biases and healthcare disparities. These findings provide a summary on multimodal data fusion as applied to health diagnosis/prognosis problems. Few papers compared the outputs of a multimodal approach with a unimodal prediction. However, those that did achieved an average increase of 6.4% in predictive accuracy. Multi-modal machine learning, while more robust in its estimations over unimodal methods, has drawbacks in its scalability and the time-consuming nature of information concatenation.

A Deep Generative-Discriminative Learning for Multimodal Representation in Imaging Genetics

Imaging genetics, one of the foremost emerging topics in the medical imaging field, analyzes the inherent relations between neuroimaging and genetic data. As deep learning has gained widespread acceptance in many applications, pioneering studies employed deep learning frameworks for imaging genetics. However, existing approaches suffer from some limitations. First, they often adopt a simple strategy for joint learning of phenotypic and genotypic features. Second, their findings have not been extended to biomedical applications, e.g., degenerative brain disease diagnosis and cognitive score prediction. Finally, existing studies perform insufficient and inappropriate analyses from the perspective of data science and neuroscience. In this work, we propose a novel deep learning framework to simultaneously tackle the aforementioned issues. Our proposed framework learns to effectively represent the neuroimaging and the genetic data jointly, and achieves state-of-the-art performance when used for Alzheimer's disease and mild cognitive impairment identification. Furthermore, unlike the existing methods, the framework enables learning the relation between imaging phenotypes and genotypes in a nonlinear way without any prior neuroscientific knowledge. To demonstrate the validity of our proposed framework, we conducted experiments on a publicly available dataset and analyzed the results from diverse perspectives. Based on our experimental results, we believe that the proposed framework has immense potential to provide new insights and perspectives in deep learning-based imaging genetics studies.

Orthogonalized Kernel Debiased Machine Learning for Multimodal Data Analysis

Multimodal imaging has transformed neuroscience research. While it presents unprecedented opportunities, it also imposes serious challenges. Particularly, it is difficult to combine the merits of the interpretability attributed to a simple association model with the flexibility achieved by a highly adaptive nonlinear model. In this article, we propose an orthogonalized kernel debiased machine learning approach, which is built upon the Neyman orthogonality and a form of decomposition orthogonality, for multimodal data analysis. We target the setting that naturally arises in almost all multimodal studies, where there is a primary modality of interest, plus additional auxiliary modalities. We establish the root-N-consistency and asymptotic normality of the estimated primary parameter, the semi-parametric estimation efficiency, and the asymptotic validity of the confidence band of the predicted primary modality effect. Our proposal enjoys, to a good extent, both model interpretability and model flexibility. It is also considerably different from the existing statistical methods for multimodal data integration, as well as the orthogonality-based methods for high-dimensional inferences. We demonstrate the efficacy of our method through both simulations and an application to a multimodal neuroimaging study of Alzheimer's disease.

Analysis of the superior temporal gyrus as a possible biomarker in schizophrenia using voxel-based morphometry of the brain magnetic resonance imaging: a comprehensive review

The lack of predictive biomarkers for therapeutic responses to schizophrenia leads clinical procedures to be decided without taking into account the subjects' neuroanatomical features, a consideration, which could help in identifying specific pharmacological treatments for the remission of symptoms. Magnetic resonance imaging (MRI) is a technique widely used for radiological diagnosis and produces 3-dimensional images in excellent anatomical detail, and with a great capacity to differentiate soft tissue. Various MRI techniques of the human brain have emerged as a result of research, enabling structural tests that may help to in consolidate previous findings and lead to the discovery of new patterns of abnormality in schizophrenia. A literature review was undertaken to assess the superior temporal gyrus (STG) as a possible biomarker in schizophrenia with the use of voxel-based morphometry of the brain using MRI. Many findings in studies of schizophrenia using MRI have been inconclusive and, in some cases, conflicting, although interesting results have been obtained when attempting to correlate neuroimaging changes with aspects of clinical features and prognosis of the disease. The individuals affected by this mental illness appear to have smaller STG volumes when compared to healthy controls and also to subjects with a diagnosis of first-episode affective psychosis or groups of individuals at high risk of psychosis. However, the wide variety of definitions surrounding the STG found in a number of studies is a contributing factor to the lack of correlation between brain abnormalities and clinical symptoms. For instance, disagreements have arisen due to studies using regions of interest to analyze the STG whereas other studies prioritize the analysis of only STG subregions or specific supratemporal plane regions. It is necessary to standardize the nomenclature of the areas to be studied in the future, as this will enable more consistent results, allowing higher clinical and morphological correlations.

A generalized kernel machine approach to identify higher-order composite effects in multi-view datasets, with application to adolescent brain development and osteoporosis

In recent years, a comprehensive study of complex disease with multi-view datasets (e.g., multi-omics and imaging scans) has been a focus and forefront in biomedical research. State-of-the-art biomedical technologies are enabling us to collect multi-view biomedical datasets for the study of complex diseases. While all the views of data tend to explore complementary information of disease, analysis of multi-view data with complex interactions is challenging for a deeper and holistic understanding of biological systems. In this paper, we propose a novel generalized kernel machine approach to identify higher-order composite effects in multi-view biomedical datasets (GKMAHCE). This generalized semi-parametric (a mixed-effect linear model) approach includes the marginal and joint Hadamard product of features from different views of data. The proposed kernel machine approach considers multi-view data as predictor variables to allow a more thorough and comprehensive modeling of a complex trait. We applied GKMAHCE approach to both synthesized datasets and real multi-view datasets from adolescent brain development and osteoporosis study. Our experiments demonstrate that the proposed method can effectively identify higher-order composite effects and suggest that corresponding features (genes, region of interests, and chemical taxonomies) function in a concerted effort. We show that the proposed method is more generalizable than existing ones. To promote reproducible research, the source code of the proposed method is available at.

Pathogenic Factors Identification of Brain Imaging and Gene in Late Mild Cognitive Impairment

Mild cognitive impairment (MCI) is a dangerous signal of severe cognitive decline. It can be separated into two steps: early MCI (EMCI) and late MCI (LMCI). As the post-state of MCI and pre-state of Alzheimer's disease (AD), LMCI receives insufficient attention in the field of brain science, causing the internal mechanism of LMCI has not been well understood. To better explore the focus and pathological mechanism of LMCI, a method called genetic evolved random forest (GERF) is applied. Resting functional magnetic resonance imaging (rfMRI) and gene data are obtained from 62 subjects (36 LMCI and 26 normal controls), and Pearson correlation analysis is adopted to perform the multimodal fusion of two types of data to construct fusion features. We identified pathogenic brain regions and genes that are highly related to LMCI using GERF and achieves a good effect. Compared with the normal control (NC) group, the abnormal brain regions of LMCI are PUT.L, PreCG.L, IFGtriang.R, REC.R, DCG.R, PoCG.L, and HES.L, and the pathogenic genes are FHIT, RF00019, FRMD4A, PTPRD, and RBFOX1. More importantly, most of these risk genes and abnormal brain regions have been confirmed to be related to AD and MCI in previous studies. In this study, we mapped them to LMCI with higher accuracies, so as to provide a more robust understanding of the physiological mechanism of MCI.

IRC-Fuse: improved and robust prediction of redox-sensitive cysteine by fusing of multiple feature representations

Redox-sensitive cysteine (RSC) thiol contributes to many biological processes. The identification of RSC plays an important role in clarifying some mechanisms of redox-sensitive factors; nonetheless, experimental investigation of RSCs is expensive and time-consuming. The computational approaches that quickly and accurately identify candidate RSCs using the sequence information are urgently needed. Herein, an improved and robust computational predictor named IRC-Fuse was developed to identify the RSC by fusing of multiple feature representations. To enhance the performance of our model, we integrated the probability scores evaluated by the random forest models implementing different encoding schemes. Cross-validation results exhibited that the IRC-Fuse achieved accuracy and AUC of 0.741 and 0.807, respectively. The IRC-Fuse outperformed exiting methods with improvement of 10% and 13% on accuracy and MCC, respectively, over independent test data. Comparative analysis suggested that the IRC-Fuse was more effective and promising than the existing predictors. For the convenience of experimental scientists, the IRC-Fuse online web server was implemented and publicly accessible at http://kurata14.bio.kyutech.ac.jp/IRC-Fuse/ .

Robust kernel canonical correlation analysis to detect gene-gene co-associations: A case study in genetics

The kernel canonical correlation analysis based U-statistic (KCCU) is being used to detect nonlinear gene-gene co-associations. Estimating the variance of the KCCU is however computationally intensive. In addition, the kernel canonical correlation analysis (kernel CCA) is not robust to contaminated data. Using a robust kernel mean element and a robust kernel (cross)-covariance operator potentially enables the use of a robust kernel CCA, which is studied in this paper. We first propose an influence function-based estimator for the variance of the KCCU. We then present a non-parametric robust KCCU, which is designed for dealing with contaminated data. The robust KCCU is less sensitive to noise than KCCU. We investigate the proposed method using both synthesized and real data from the Mind Clinical Imaging Consortium (MCIC). We show through simulation studies that the power of the proposed methods is a monotonically increasing function of sample size, and the robust test statistics bring incremental gains in power. To demonstrate the advantage of the robust kernel CCA, we study MCIC data among 22,442 candidate Schizophrenia genes for gene-gene co-associations. We select 768 genes with strong evidence for shedding light on gene-gene interaction networks for Schizophrenia. By performing gene ontology enrichment analysis, pathway analysis, gene-gene network and other studies, the proposed robust methods can find undiscovered genes in addition to significant gene pairs, and demonstrate superior performance over several of current approaches.

Brain Imaging Genomics: Integrated Analysis and Machine Learning

Brain imaging genomics is an emerging data science field, where integrated analysis of brain imaging and genomics data, often combined with other biomarker, clinical and environmental data, is performed to gain new insights into the phenotypic, genetic and molecular characteristics of the brain as well as their impact on normal and disordered brain function and behavior. It has enormous potential to contribute significantly to biomedical discoveries in brain science. Given the increasingly important role of statistical and machine learning in biomedicine and rapidly growing literature in brain imaging genomics, we provide an up-to-date and comprehensive review of statistical and machine learning methods for brain imaging genomics, as well as a practical discussion on method selection for various biomedical applications.