Dataset Information

Hierarchical Clustering Analyses of Plasma Proteins in Subjects With Cardiovascular Risk Factors Identify Informative Subsets Based on Differential Levels of Angiogenic and Inflammatory Biomarkers.

ABSTRACT: Agglomerative hierarchical clustering analysis (HCA) is a commonly used unsupervised machine learning approach for identifying informative natural clusters of observations. HCA is performed by calculating a pairwise dissimilarity matrix and then clustering similar observations until all observations are grouped within a cluster. Verifying the empirical clusters produced by HCA is complex and not well studied in biomedical applications. Here, we demonstrate the comparability of a novel HCA technique with one that was used in previous biomedical applications while applying both techniques to plasma angiogenic (FGF, FLT, PIGF, Tie-2, VEGF, VEGF-D) and inflammatory (MMP1, MMP3, MMP9, IL8, TNFα) protein data to identify informative subsets of individuals. Study subjects were diagnosed with mild cognitive impairment due to cerebrovascular disease (MCI-CVD). Through comparison of the two HCA techniques, we were able to identify subsets of individuals, based on differences in VEGF (p < 0.001), MMP1 (p < 0.001), and IL8 (p < 0.001) levels. These profiles provide novel insights into angiogenic and inflammatory pathologies that may contribute to VCID.

SUBMITTER: Winder Z

PROVIDER: S-EPMC7016016 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Hierarchical Clustering Analyses of Plasma Proteins in Subjects With Cardiovascular Risk Factors Identify Informative Subsets Based on Differential Levels of Angiogenic and Inflammatory Biomarkers.

Winder Zachary Z Sudduth Tiffany L TL Fardo David D Cheng Qiang Q Goldstein Larry B LB Nelson Peter T PT Schmitt Frederick A FA Jicha Gregory A GA Wilcock Donna M DM

Frontiers in neuroscience 20200206

Agglomerative hierarchical clustering analysis (HCA) is a commonly used unsupervised machine learning approach for identifying informative natural clusters of observations. HCA is performed by calculating a pairwise dissimilarity matrix and then clustering similar observations until all observations are grouped within a cluster. Verifying the empirical clusters produced by HCA is complex and not well studied in biomedical applications. Here, we demonstrate the comparability of a novel HCA techni ...[more]

PMID: 32116527

Similar Datasets

Project description:IntroductionAnti-Ro52/tripartite motif-containing protein 21 (TRIM21) IgG is one of the most common autoantibodies found in systemic autoimmune diseases and is typically found in conjunction with anti-Ro60 and/or anti-La. A retrospective, cross-sectional study was undertaken to examine the clinical and laboratory features of two serological subsets: patients with anti-Ro52/TRIM21 autoantibodies in the absence of anti-Ro60 and anti-La (isolated anti-Ro52/TRIM21) and patients with anti-Ro52/TRIM21 in the presence of anti-Ro60 and/or anti-La.MethodsOver a 12-month period, patients tested positive for anti-Ro52/TRIM21 via line immunoassay (LIA) at the Westmead Hospital (Australia) immunopathology laboratory were included. The presence of anti-Ro60 and/or anti-La via same LIA was noted. Associated laboratory and medical records were perused to extract demographic, laboratory, and clinical information.ResultsThere were 346 patients within the study period, and 39.9% of the patients positive for anti-Ro52/TRIM21 lacked anti-Ro60/anti-La autoantibodies. Isolated anti-Ro52/TRIM21 patients tend to be older, have lower anti-Ro52/TRIM21 titres, have lower rheumatoid factors, and have lower proportions of neutropaenia compared to patients who were positive for anti-Ro52/TRIM21 and anti-Ro60/La. This occurred independent to diagnoses of Sjögren's syndrome or systemic lupus erythematosus. Coexisting neurological syndromes, pulmonary pathologies, and malignancies were more prevalent in the isolated anti-Ro52/TRIM21 subset.ConclusionsPatients with isolated anti-Ro52/TRIM21 tend to have distinct and important clinical and laboratory associations. It is unclear if these patients evolve or remain a stable subset and how they originate immunologically. Longitudinal and prospective studies are required to ascertain the overall predictive and prognostic value of this stratification. Key Points • Anti-Ro52/TRIM21 is an autoantibody found in autoimmunity and non-immunological conditions. • Sixty percent of anti-Ro52/TRIM21 patients are positive for anti-Ro60. • Isolated anti-Ro52/TRIM21 has reduced anti-Ro52/TRIM21 and rheumatoid factor titres. • Isolated anti-Ro52/TRIM21 is associated with anaemia and malignancies.

Project description:Cystinosis is a rare, devastating hereditary disease secondary to recessive CTNS gene mutations. The most commonly used diagnostic method is confirmation of an elevated leukocyte cystine level; however, this method is expensive and difficult to perform. This study aimed to identify candidate biomarkers for the diagnosis and follow-up of cystinosis based on multiomics studies. The study included three groups: newly-diagnosed cystinosis patients (patient group, n = 14); cystinosis patients under treatment (treatment group, n = 19); and healthy controls (control group, n = 30). Plasma metabolomics analysis identified 10 metabolites as candidate biomarkers that differed between the patient and control groups [L-serine, taurine, lyxose, 4-trimethylammoniobutanoic acid, orotic acid, glutathione, PE(O-18:1(9Z)/0:0), 2-hydroxyphenyl acetic acid, acetyl-N-formil-5-metoxikinuramine, 3-indoxyl sulphate]. As compared to the healthy control group, in the treatment group, hypotaurine, phosphatidylethanolamine, N-acetyl-d-mannosamine, 3-indolacetic acid, p-cresol, phenylethylamine, 5-aminovaleric acid, glycine, creatinine, and saccharic acid levels were significantly higher, and the metabolites quinic acid, capric acid, lenticin, xanthotoxin, glucose-6-phosphate, taurine, uric acid, glyceric acid, alpha-D-glucosamine phosphate, and serine levels were significantly lower. Urinary metabolomic analysis clearly differentiated the patient group from the control group by means of higher allo-inositol, talose, glucose, 2-hydroxybutiric acid, cystine, pyruvic acid, valine, and phenylalanine levels, and lower metabolite (N-acetyl-L-glutamic acid, 3-aminopropionitrile, ribitol, hydroquinone, glucuronic acid, 3-phosphoglycerate, xanthine, creatinine, and 5-aminovaleric acid) levels in the patient group. Urine metabolites were also found to be significantly different in the treatment group than in the control group. Thus, this study identified candidate biomarkers that could be used for the diagnosis and follow-up of cystinosis.

Project description:Breast cancer includes several subtypes with distinct characteristic biological, pathologic, and clinical features. Elucidating subtype-specific genetic etiology could provide insights into the heterogeneity of breast cancer to facilitate the development of improved prevention and treatment approaches. In this study, we conducted pairwise case-case comparisons among five breast cancer subtypes by applying a case-case genome-wide association study (CC-GWAS) approach to summary statistics data of the Breast Cancer Association Consortium. The approach identified 13 statistically significant loci and eight suggestive loci, the majority of which were identified from comparisons between triple-negative breast cancer (TNBC) and luminal A breast cancer. Associations of lead variants in 12 loci remained statistically significant after accounting for previously reported breast cancer susceptibility variants, among which, two were genome-wide significant. Fine mapping implicated putative functional/causal variants and risk genes at several loci, e.g., 3q26.31/TNFSF10, 8q22.3/NACAP1/GRHL2, and 8q23.3/LINC00536/TRPS1, for TNBC as compared with luminal cancer. Functional investigation further identified rs16867605 at 8q22.3 as a SNP that modulates the enhancer activity of GRHL2. Subtype-informative polygenic risk scores (PRS) were derived, and patients with a high subtype-informative PRS had an up to two-fold increased risk of being diagnosed with TNBC instead of luminal cancers. The CC-GWAS PRS remained statistically significant after adjusting for TNBC PRS derived from traditional case-control GWAS in The Cancer Genome Atlas and the African Ancestry Breast Cancer Genetic Consortium. The CC-GWAS PRS was also associated with overall survival and disease-specific survival among patients with breast cancer. Overall, these findings have advanced our understanding of the genetic etiology of breast cancer subtypes, particularly for TNBC. Significance: The discovery of subtype-informative genetic risk variants for breast cancer advances our understanding of the etiologic heterogeneity of breast cancer, which could accelerate the identification of targets and personalized strategies for prevention and treatment.

Dataset Information

Hierarchical Clustering Analyses of Plasma Proteins in Subjects With Cardiovascular Risk Factors Identify Informative Subsets Based on Differential Levels of Angiogenic and Inflammatory Biomarkers.

Publications

Hierarchical Clustering Analyses of Plasma Proteins in Subjects With Cardiovascular Risk Factors Identify Informative Subsets Based on Differential Levels of Angiogenic and Inflammatory Biomarkers.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets