Background: There is a significant need for rapid and cost-effective biomarkers of Alzheimer’s disease (AD) for advancement of clinical practice and therapeutic trials. Objective: The aim of the current study was to cross-validate our previously published serum-based algorithm on an independent assay platform as well as validate across tissues and species. Preliminary analyses were conducted to examine the utility in distinguishing AD from non-AD neurological disease (Parkinson’s disease, PD). Methods: Serum proteins from our previously published algorithm were quantified from 150 AD cases and 150 controls on the Meso Scale Discovery (MSD) platform. Serum samples were analyzed from 49 PD cases and compared to a random sample of 51 AD cases and 62 controls. Support vector machines (SVM) were used to discriminate PD versus AD versus controls. Human and AD mouse model microvessel images were quantified with HAMAMATSU imaging software. Mouse serum biomarkers were assayed via MSD. Results: Analysis of 21 serum proteins from 150 AD cases and 150 controls yielded an algorithm with sensitivity and specificity of 0.90 for correctly classifying AD. This multi-marker approach was then validated across species and tissue. Assay of the top proteins in human and AD mouse model brain microvessels correctly classified 90–100% of the samples. SVM analyses were highly accurate at distinguishing PD versus AD versus controls. Conclusions: This serum-based biomarker panel should be tested in a community-based setting to determine its utility as a first-line screen for AD and non-AD neurological diseases for primary care providers.

Background: Breast cancer is worldwide the second most common type of cancer after lung cancer. Traditional mammography and Tissue Microarray has been studied for early cancer detection and cancer prediction. However, there is a need for more reliable diagnostic tools for early detection of breast cancer. This can be a challenge due to a number of factors and logistics. First, obtaining tissue biopsies can be difficult. Second, mammography may not detect small tumors, and is often unsatisfactory for younger women who typically have dense breast tissue. Lastly, breast cancer is not a single homogeneous disease but consists of multiple disease states, each arising from a distinct molecular mechanism and having a distinct clinical progression path which makes the disease difficult to detect and predict in early stages. Results: In the paper, we present a Support Vector Machine based on Recursive Feature Elimination and Cross Validation (SVM-RFE-CV) algorithm for early detection of breast cancer in peripheral blood and show how to use SVM-RFE-CV to model the classification and prediction problem of early detection of breast cancer in peripheral blood. The training set which consists of 32 health and 33 cancer samples and the testing set consisting of 31 health and 34 cancer samples were randomly separated from a dataset of peripheral blood of breast cancer that is downloaded from Gene Express Omnibus. First, we identified the 42 differentially expressed biomarkers between "normal" and "cancer". Then, with the SVM-RFE-CV we extracted 15 biomarkers that yield zero cross validation score. Lastly, we compared the classification and prediction performance of SVM-RFE-CV with that of SVM and SVM Recursive Feature Elimination (SVM-RFE). Conclusions: We found that 1) the SVM-RFE-CV is suitable for analyzing noisy high-throughput microarray data, 2) it outperforms SVM-RFE in the robustness to noise and in the ability to recover informative features, and 3) it can improve the prediction performance (Area Under Curve) in the testing data set from 0.5826 to 0.7879. Further pathway analysis showed that the biomarkers are associated with Signaling, Hemostasis, Hormones, and Immune System, which are consistent with previous findings. Our prediction model can serve as a general model for biomarker discovery in early detection of other cancers. In the future, Polymerase Chain Reaction (PCR) is planned for validation of the ability of these potential biomarkers for early detection of breast cancer.

Next-Generation Sequencing (NGS) technologies and Genome-Wide Association Studies (GWAS) generate millions of reads and hundreds of datasets, and there is an urgent need for a better way to accurately interpret and distill such large amounts of data. Extensive pathway and network analysis allow for the discovery of highly significant pathways from a set of disease vs. healthy samples in the NGS and GWAS. Knowledge of activation of these processes will lead to elucidation of the complex biological pathways affected by drug treatment, to patient stratification studies of new and existing drug treatments, and to understanding the underlying anti-cancer drug effects. There are approximately 141 biological human pathway resources as of Jan 2012 according to the Pathguide database. However, most currently available resources do not contain disease, drug or organ specificity information such as disease-pathway, drug-pathway, and organ-pathway associations. Systematically integrating pathway, disease, drug and organ specificity together becomes increasingly crucial for understanding the interrelationships between signaling, metabolic and regulatory pathway, drug action, disease susceptibility, and organ specificity from high-throughput omics data (genomics, transcriptomics, proteomics and metabolomics).

Early detection of breast cancer in blood is both appealing clinically and challenging technically due to the disease's illusive nature and heterogeneity. Today, even though major breast cancer subtypes have been characterized, i.e., luminal A, luminal B, HER2+, and basal-like, little is known about the heterogeneity of breast cancer in blood, which could help to discover minimally invasive protein biomarkers with which clinical researchers can detect, classify, and monitor different breast cancer subtypes.

Accurate detection is still a challenge in machine learning (ML) for Alzheimer’s disease (AD). Class imbalance in imbalanced AD data is another big challenge for machine-learning algorithms working under the assumption that the data are evenly distributed within classes. Here, we present a hyperparameter tuning workflow with high-performance computing (HPC) for imbalanced data related to prevalent mild cognitive impairment (MCI) and AD in the Health and Aging Brain Study-Health Disparities (HABS-HD) project. We applied a single-node multicore parallel mode to hyperparameter tuning of gamma, cost, and class weight using a support vector machine (SVM) model with 10 times repeated fivefold cross-validation. We executed the hyperparameter tuning workflow with R’s bigmemory, foreach, and doParallel packages on Texas Advanced Computing Center (TACC)’s Lonestar6 system. The computational time was dramatically reduced by up to 98.2% for the high-performance SVM hyperparameter tuning model, and the performance of cross-validation was also improved (the positive predictive value and the negative predictive value at base rate 12% were, respectively, 16.42% and 92.72%). Our results show that a single-node multicore parallel structure and high-performance SVM hyperparameter tuning model can deliver efficient and fast computation and achieve outstanding agility, simplicity, and productivity for imbalanced data in AD applications.

In the biopharmaceutical industry, biomarkers define molecular taxonomies of patients and diseases and serve as surrogate endpoints in early-phase drug trials. Molecular biomarkers can be much more sensitive than traditional lab tests. Discriminating disease biomarkers by traditional method such as DNA microarray has proved challenging. Alternative splicing isoform represents a new class of diagnostic biomarkers. Recent scientific evidence is demonstrating that the differentiation and quantification of individual alternative splicing isoforms could improve insights into disease diagnosis and management. Identifying and characterizing alternative splicing isoforms are essential to the study of molecular mechanisms and early detection of complex diseases such as breast cancer. However, there are limitations with traditional methods used for alternative splicing isoform determination such as transcriptome-level, low level of coverage and poor focus on alternative splicing.