Ethnic differences in age-related changes of postpartum hemoglobin levels

Unsupervised analysis of classical biomedical markers: robustness and medical relevance of patient clustering using bioinformatics tools

Motivation

It has been proposed that clustering clinical markers, such as blood test results, can be used to stratify patients. However, the robustness of clusters formed with this approach to data pre-processing and clustering algorithm choices has not been evaluated, nor has clustering reproducibility. Here, we made use of the NHANES survey to compare clusters generated with various combinations of pre-processing and clustering algorithms, and tested their reproducibility in two separate samples.

Method

Values of 44 biomarkers and 19 health/life style traits were extracted from the National Health and Nutrition Examination Survey (NHANES). The 1999–2002 survey was used for training, while data from the 2003–2006 survey was tested as a validation set. Twelve combinations of pre-processing and clustering algorithms were applied to the training set. The quality of the resulting clusters was evaluated both by considering their properties and by comparative enrichment analysis. Cluster assignments were projected to the validation set (using an artificial neural network) and enrichment in health/life style traits in the resulting clusters was compared to the clusters generated from the original training set.

Results

The clusters obtained with different pre-processing and clustering combinations differed both in terms of cluster quality measures and in terms of reproducibility of enrichment with health/life style properties. Z-score normalization, for example, dramatically improved cluster quality and enrichments, as compared to unprocessed data, regardless of the clustering algorithm used. Clustering diabetes patients revealed a group of patients enriched with retinopathies. This could indicate that routine laboratory tests can be used to detect patients suffering from complications of diabetes, although other explanations for this observation should also be considered.

Conclusions

Clustering according to classical clinical biomarkers is a robust process, which may help in patient stratification. However, optimization of the pre-processing and clustering process may be still required.

Towards an Age-Phenome Knowledge-base

Background

Currently, data about age-phenotype associations are not systematically organized and cannot be studied methodically. Searching for scientific articles describing phenotypic changes reported as occurring at a given age is not possible for most ages.

Results

Here we present the Age-Phenome Knowledge-base (APK), in which knowledge about age-related phenotypic patterns and events can be modeled and stored for retrieval. The APK contains evidence connecting specific ages or age groups with phenotypes, such as disease and clinical traits. Using a simple text mining tool developed for this purpose, we extracted instances of age-phenotype associations from journal abstracts related to non-insulin-dependent Diabetes Mellitus. In addition, links between age and phenotype were extracted from clinical data obtained from the NHANES III survey. The knowledge stored in the APK is made available for the relevant research community in the form of 'Age-Cards', each card holds the collection of all the information stored in the APK about a particular age. These Age-Cards are presented in a wiki, allowing community review, amendment and contribution of additional information. In addition to the wiki interaction, complex searches can also be conducted which require the user to have some knowledge of database query construction.

Conclusions

The combination of a knowledge model based repository with community participation in the evolution and refinement of the knowledge-base makes the APK a useful and valuable environment for collecting and curating existing knowledge of the connections between age and phenotypes.