Cardiovascular disease, heart failure, chronic kidney disease and depression independently increase the risk of incident diabetes.

AIMS/HYPOTHESIS: Diabetes increases the risk of cardiovascular disease (CVD) and heart failure, as well as other serious complications, such as renal disease and depression. However, these conditions are often present prior to diabetes diagnosis. We sought to determine whether they increase the risk of developing diabetes independent of other risk factors. METHODS: We identified 58,056 non-diabetic adults aged ≥30 years with no evidence of diabetes. Using electronic medical records, we identified the presence of four conditions at baseline (CVD, heart failure, renal disease and depression) and then estimated diabetes incidence over 5 years separately for patients with and without each of these conditions. Each incidence estimate was adjusted for baseline values of age, sex, fasting glucose, body mass index, systolic blood pressure, triacylglycerol, HDL-cholesterol, smoking and the presence of the other three conditions. RESULTS: Patients with CVD were 35% (95% CI 23-48%) more likely to develop diabetes after controlling for other risk factors. Heart failure was independently associated with an increase in diabetes incidence of 48% (95% CI 27-73%), and depression was associated with a 10% (95% CI 2-20%) increase. Chronic kidney disease was associated with a non-significant risk increase of 10% (95% CI -2-25%). CONCLUSIONS/INTERPRETATION: Complications of diabetes are more prevalent among patients who will ultimately develop diabetes, and increase the risk of diabetes independently of other known risk factors. The apparent bidirectional relationships suggest that primary prevention of CVD may also help prevent diabetes.

Identifying marker genes in transcription profiling data using a mixture of feature relevance experts.

Transcription profiling experiments permit the expression levels of many genes to be measured simultaneously. Given profiling data from two types of samples, genes that most distinguish the samples (marker genes) are good candidates for subsequent in-depth experimental studies and developing decision support systems for diagnosis, prognosis, and monitoring. This work proposes a mixture of feature relevance experts as a method for identifying marker genes and illustrates the idea using published data from samples labeled as acute lymphoblastic and myeloid leukemia (ALL, AML). A feature relevance expert implements an algorithm that calculates how well a gene distinguishes samples, reorders genes according to this relevance measure, and uses a supervised learning method [here, support vector machines (SVMs)] to determine the generalization performances of different nested gene subsets. The mixture of three feature relevance experts examined implement two existing and one novel feature relevance measures. For each expert, a gene subset consisting of the top 50 genes distinguished ALL from AML samples as completely as all 7,070 genes. The 125 genes at the union of the top 50s are plausible markers for a prototype decision support system. Chromosomal aberration and other data support the prediction that the three genes at the intersection of the top 50s, cystatin C, azurocidin, and adipsin, are good targets for investigating the basic biology of ALL/AML. The same data were employed to identify markers that distinguish samples based on their labels of T cell/B cell, peripheral blood/bone marrow, and male/female. Selenoprotein W may discriminate T cells from B cells. Results from analysis of transcription profiling data from tumor/nontumor colon adenocarcinoma samples support the general utility of the aforementioned approach. Theoretical issues such as choosing SVM kernels and their parameters, training and evaluating feature relevance experts, and the impact of potentially mislabeled samples on marker identification (feature selection) are discussed.

Analysis of molecular profile data using generative and discriminative methods.

A modular framework is proposed for modeling and understanding the relationships between molecular profile data and other domain knowledge using a combination of generative (here, graphical models) and discriminative [Support Vector Machines (SVMs)] methods. As illustration, naive Bayes models, simple graphical models, and SVMs were applied to published transcription profile data for 1,988 genes in 62 colon adenocarcinoma tissue specimens labeled as tumor or nontumor. These unsupervised and supervised learning methods identified three classes or subtypes of specimens, assigned tumor or nontumor labels to new specimens and detected six potentially mislabeled specimens. The probability parameters of the three classes were utilized to develop a novel gene relevance, ranking, and selection method. SVMs trained to discriminate nontumor from tumor specimens using only the 50-200 top-ranked genes had the same or better generalization performance than the full repertoire of 1,988 genes. Approximately 90 marker genes were pinpointed for use in understanding the basic biology of colon adenocarcinoma, defining targets for therapeutic intervention and developing diagnostic tools. These potential markers highlight the importance of tissue biology in the etiology of cancer. Comparative analysis of molecular profile data is proposed as a mechanism for predicting the physiological function of genes in instances when comparative sequence analysis proves uninformative, such as with human and yeast translationally controlled tumour protein. Graphical models and SVMs hold promise as the foundations for developing decision support systems for diagnosis, prognosis, and monitoring as well as inferring biological networks.

Integrating naive Bayes models and external knowledge to examine copper and iron homeostasis in S. cerevisiae.

A novel suite of analytical techniques and visualization tools are applied to 78 published transcription profiling experiments monitoring 5,687 Saccharomyces cerevisiae genes in studies examining cell cycle, responses to stress, and diauxic shift. A naive Bayes model discovered and characterized 45 classes of gene profile vectors. An enrichment measure quantified the association between these classes and specific external knowledge defined by four sets of categories to which genes can be assigned: 106 protein functions, 5 stages of the cell cycle, 265 transcription factors, and 16 chromosomal locations. Many of the 38 genes in class 42 are known to play roles in copper and iron homeostasis. The 17 uncharacterized open reading frames in this class may be involved in similar homeostatic processes; human homologs of two of them could be associated with as yet undefined disease states arising from aberrant metal ion regulation. The Met4, Met31, and Met32 transcription factors may play a role in coregulating genes involved in copper and iron metabolism. Extensions of the simple graphical model used for clustering to learning more complex models of genetic networks are discussed.