Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

The reference human genome sequence set the stage for studies of genetic variation and its association with human disease, but epigenomic studies lack a similar reference. To address this need, the NIH Roadmap Epigenomics Consortium generated the largest collection so far of human epigenomes for primary cells and tissues. Here we describe the integrative analysis of 111 reference human epigenomes generated as part of the programme, profiled for histone modification patterns, DNA accessibility, DNA methylation and RNA expression. We establish global maps of regulatory elements, define regulatory modules of coordinated activity, and their likely activators and repressors. We show that disease- and trait-associated genetic variants are enriched in tissue-specific epigenomic marks, revealing biologically relevant cell types for diverse human traits, and providing a resource for interpreting the molecular basis of human disease. Our results demonstrate the central role of epigenomic information for understanding gene regulation, cellular differentiation and human disease.

Integrative analysis of 111 reference human epigenomes

Persistent postsurgical pain: risk factors and prevention

The brain is the key organ of the response to stress because it determines what is threatening and, therefore, potentially stressful, as well as the physiological and behavioral responses which can be either adaptive or damaging. Stress involves two-way communication between the brain and the cardiovascular, immune, and other systems via neural and endocrine mechanisms. Beyond the "flight-or-fight" response to acute stress, there are events in daily life that produce a type of chronic stress and lead over time to wear and tear on the body ("allostatic load"). Yet, hormones associated with stress protect the body in the short-run and promote adaptation ("allostasis"). The brain is a target of stress, and the hippocampus was the first brain region, besides the hypothalamus, to be recognized as a target of glucocorticoids. Stress and stress hormones produce both adaptive and maladaptive effects on this brain region throughout the life course. Early life events influence life-long patterns of emotionality and stress responsiveness and alter the rate of brain and body aging. The hippocampus, amygdala, and prefrontal cortex undergo stress-induced structural remodeling, which alters behavioral and physiological responses. As an adjunct to pharmaceutical therapy, social and behavioral interventions such as regular physical activity and social support reduce the chronic stress burden and benefit brain and body health and resilience.

/pdf/physiology-and-neurobiology-of-stress-and-adaptation-central-2xtqi7pnfe.pdf

Physiology and Neurobiology of Stress and Adaptation: Central Role of the Brain

A neurotrophic model for stress-related mood disorders.

Epigenome-wide association studies (EWAS) of heterogenous blood cells have identified CpG sites associated with chronic HIV infection, which offer limited knowledge of cell-type specific methylation patterns associated with HIV infection. Applying a computational deconvolution method validated by capture bisulfite DNA methylation sequencing, we conducted a cell type-based EWAS and identified differentially methylated CpG sites specific for chronic HIV infection among five immune cell types in blood: CD4+ T-cells, CD8+ T-cells, B cells, Natural Killer (NK) cells, and monocytes in two independent cohorts (Ntotal=1,134). Differentially methylated CpG sites for HIV-infection were highly concordant between the two cohorts. Cell-type level meta-EWAS revealed distinct patterns of HIV-associated differential CpG methylation, where 67% of CpG sites were unique to individual cell types (false discovery rate, FDR <0.05). CD4+ T-cells had the largest number of HIV-associated CpG sites (N=1,472) compared to any other cell type. Genes harboring statistically significant CpG sites are involved in immunity and HIV pathogenesis (e.g. CX3CR1 in CD4+ T-cells, CCR7 in B cells, IL12R in NK cells, LCK in monocytes). More importantly, HIV-associated CpG sites were overrepresented for hallmark genes involved in cancer pathology (FDR<0.05) (e.g. BCL family, PRDM16, PDCD1LGD, ESR1, DNMT3A, NOTCH2). HIV-associated CpG sites were enriched among genes involved in HIV pathogenesis and oncogenesis such as Kras-signaling, interferon-α and −γ, TNF-α, inflammatory, and apoptotic pathways. Our findings are novel, uncovering cell-type specific modifications in the host epigenome for people with HIV that contribute to the growing body of evidence regarding pathogen-induced epigenetic oncogenicity, specifically on HIV and its comorbidity with cancers.

/pdf/cell-type-specific-ewas-identifies-genes-involved-in-hiv-2t57bz8t.pdf

Cell-type specific EWAS identifies genes involved in HIV pathogenesis and oncogenesis among people with HIV infection

Motivation High-quality imaging analyses have been proposed to drive innovation in biomedical and biological research. However, the application of images remains underexploited because of the limited capacity of human vision and the challenges in extracting quantitative information from images. Computationally extracting quantitative information from images is critical to overcoming this limitation. Here, we present a novel R package, i2d, to simulate data from an image based on digital convolution. Results The R package i2d allows users to transform an image into a simulated dataset that can be used to extract and analyze complex information in biomedical and biological research. The package also includes three novel and efficient methods for graph clustering based on simulated data, which can be used to dissect complex gene networks into sub-clusters that have similar biological functions. Availability and implementation The code, the documentation, a tutorial and example data are available on an open source at: github.com/XiaoyuLiang/i2d. Supplementary information Supplementary data are available at Bioinformatics online.

/pdf/i2d-an-r-package-for-simulating-data-from-images-and-the-2uyz40lp21.pdf

i2d: an R package for simulating data from images and the implications in biomedical research.


 Background

Oral squamous cell carcinoma (OSCC) has poor survival rates. There is a pressing need to develop more precise risk assessment methods to tailor clinical treatment. Epigenome-wide association studies in OSCC have not produced a viable biomarker. These studies have relied on methylation array platforms, which are limited in their ability to profile the methylome. In this study, we use MethylCap-Seq (MC-Seq), a comprehensive methylation quantification technique, and brush swab samples, to develop a noninvasive, readily translatable approach to profile the methylome in OSCC patients.
Methods

Three OSCC patients underwent collection of cancer and contralateral normal tissue and brush swab biopsies, totaling 4 samples for each patient. Epigenome-wide DNA methylation quantification was performed using the SureSelectXT Methyl-Seq platform. DNA quality and methylation site resolution were compared between brush swab and tissue samples. Correlation and methylation value difference were determined for brush swabs vs. tissues for each respective patient and site (i.e., cancer or normal). Correlations were calculated between cancer and normal tissues and brush swab samples for each patient to determine the robustness of DNA methylation marks using brush swabs in clinical biomarker studies.
Results

There were no significant differences in DNA yield between tissue and brush swab samples. Mapping efficiency exceeded 90% across all samples, with no differences between tissue and brush swabs. The average number of CpG sites with at least 10x depth of coverage was 2,716,674 for brush swabs and 2,903,261 for tissues. Matched tissue and brush swabs had excellent correlation (r = 0.913 for cancer samples and r = 0.951 for normal samples). The methylation profile of the top 1,000 CpGs was significantly different between cancer and normal samples (mean p-value = 0.00021) but not different between tissues and brush swabs (mean p-value = 0.11).
Conclusions

Our results demonstrate that MC-Seq is an efficient platform for epigenome profiling in cancer biomarker studies, with broader methylome coverage than array-based platforms. Brush swab biopsy provides adequate DNA yield for MC-Seq, and taken together, our findings set the stage for development of a non-invasive methylome quantification technique for oral cancer with high translational potential.

/pdf/brush-swab-as-a-noninvasive-surrogate-for-tissue-biopsies-in-40p781znw1.pdf

Ke Xu

Papers

Cell-type specific EWAS identifies genes involved in HIV pathogenesis and oncogenesis among people with HIV infection

i2d: an R package for simulating data from images and the implications in biomedical research.

Brush Swab As A Noninvasive Surrogate for Tissue Biopsies in Oral Cancer Patients To Develop Clinically Translatable Epigenetic Biomarkers

Greater Stress and Trauma Mediate Race Differences in Epigenetic Age between Black and White Healthy Young Adults

P75. Stress and Trauma Mediate Some, But Not All, Differences in GrimAge Acceleration Between Black and White Participants in Observed Cohort Study