comoR: a software for disease comorbidity risk assessment

Abstract

BackgroundThe diagnosis of comorbidities, which refers to the coexistence of different acute and chronic diseases, is difficult due to the modern extreme specialisation of physicians. We envisage that a software dedicated to comorbidity diagnosis could result in an effective aid to the health practice.ResultsWe have developed an R software to compute novel estimators of the disease comorbidity associations. Starting from an initial diagnosis, genetic and clinical data of a patient the software identifies the risk of disease comorbidity. Then it provides a pipeline with different causal inference packages (e.g. pcalg, qtlnet etc) to predict the causal relationship of diseases. It also provides a pipeline with network regression and survival analysis tools (e.g. Net-Cox, rbsurv etc) to predict more accurate survival probability of patients. The input of this software is the initial diagnosis for a patient and the output provides evidences of disease comorbidity mapping.ConclusionsThe functions of the offer flexibility for diagnostic applications to predict disease comorbidities, and can be easily integrated to high–throughput and clinical data analysis pipelines.

Bio-Inspired ICT for Big Data Management in Healthcare

“…A comorbidity relation between two diseases occurs when the two diseases appear simultaneously in a patient, not by chance alone [4]. It is worthless to underline that some infections or diseases can coexist in the same patient only by coincidence, but there is not any pathological association between them [5]. In most cases, however, multiple diseases (chronicle or acute events) occur together in the same patient because of the associations between them.…”

“…In the biomedical research field, for example, one of the main problems is to find out the correlations between human diseases and the subtle mechanisms, at the molecular level, shared by apparently dissimilar diseases. By exploring genes-diseases relations, the pathogenesis may emerge, and it could lead to better diagnosis and treatment [5]. Diseases are more likely comorbid if they share associated genes [39].…”

“…Social dimension and social relationships can act as a selection bias in choosing one treatment plane or drug rather than others, or still shedding light on some side-effects of some drugs inhibiting the usage, or mimicking "good" behaviors of some friends in the social network. To quantify the comorbidity, the authors of [5] have used two measures: the relative risk, that is the fraction between the number of patients diagnosed with both diseases and random expectation based on disease prevalence as the quantified measure of comorbidity tendency of two disease pairs; the φ-correlation (Pearson correlation for binary variables) to measure the robustness of the comorbidity association. Beyond these measures, we need to use also a social measure of contagion, considering a dynamic evolution of comorbidities which depends on a different response of the patient linked to a genotypic-phenotypic diversity, which may induce different social behaviors and, as a consequence, to different maps (Figs.…”

“…In the biomedical research field, for example, one of the main problems is to find out the correlations between human diseases and the subtle mechanisms, at the molecular level, shared by apparently dissimilar diseases. By exploring genes-diseases relations, the pathogenesis may emerge, and it could lead to better diagnosis and treatment [5]. In this field, the study of comorbidity relations becomes crucial, as it allows to find out dynamic patterns, correlations and causality mechanisms between diseases and patients and, even more interesting, it becomes crucial to study the social dimension of comorbidity, focusing on how the strength of the ties, the connectedness, and communications lead to an evolutionary dynamics of the social network and represent a further degree of analysis in the comorbidity assessment.…”

A system biology approach reveals cellular pathway differences between follicular thyroid carcinoma and follicular thyroid adenoma

“…For such pathway studies our starting point is the differential expression of genes (DEGs) between these lesions. We can then use analytic tools to study these DEGs and discover pathways, and pathway hub genes that may affect cell functions and thereby underlie their tumour morbidity, growth and invasion [28,14,30]. The repertoire of candidate pathway factors can be extended using approaches such as gene ontology analysis and protein-protein interactions (PPI) studies.…”