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

Markovich Gordon M, Moser AM, Rubin E - PLoS ONE (2012)

Bottom Line: It has been proposed that clustering clinical markers, such as blood test results, can be used to stratify patients.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.However, optimization of the pre-processing and clustering process may be still required.

View Article: PubMed Central - PubMed

Affiliation: Sharaga Segal Dept of Microbiology and Immunology, Ben Gurion University of the Negev, Beersheba, Israel.

ABSTRACT

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.

Show MeSH

Related in: MedlinePlus

Selected clusters from the NHANES training set.(A) The mean and standard deviation of biomarker values are shown for three selected clusters generated with 4152 males 20 years of age or older from the NHANES training set, using the Z-score normalized/CLICK pre-processing/clustering combination. For each cluster, the total number of individuals (top, right) and selected health/lifestyle traits that are significantly enriched in that cluster (top, left) are provided. For each enriched term, the enrichment factor (i.e. the frequency of the term in a cluster divided by its frequency in the entire dataset) is also provided. (B) Comparison of the original values of selected biomarkers in clusters 1, 3 and 14. The values for Hb are enlarged in the top middle section of the figure.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3293863&req=5

pone-0029578-g002: Selected clusters from the NHANES training set.(A) The mean and standard deviation of biomarker values are shown for three selected clusters generated with 4152 males 20 years of age or older from the NHANES training set, using the Z-score normalized/CLICK pre-processing/clustering combination. For each cluster, the total number of individuals (top, right) and selected health/lifestyle traits that are significantly enriched in that cluster (top, left) are provided. For each enriched term, the enrichment factor (i.e. the frequency of the term in a cluster divided by its frequency in the entire dataset) is also provided. (B) Comparison of the original values of selected biomarkers in clusters 1, 3 and 14. The values for Hb are enlarged in the top middle section of the figure.

Mentions: To test whether the observed clusters can define biomedically homogenous sub-populations, we inspected other cluster characteristics. We hypothesized that if individuals in the same cluster share similar properties, then the enrichment we observed may correspond to a specific pattern of biomarker values in that cluster. To demonstrate that such correspondence can be observed, the mean biomarker values of three clusters were analyzed in the context of their health/lifestyle traits enrichment (Figure 2). Cluster 14 is highly enriched with diabetic individuals. As expected, members of this cluster are characterized by high levels of glucose, blood osmolarity and triglycerides, traits that are indeed hallmarks of unbalanced diabetic patients [7]. Cluster 1 is enriched with individuals suffering from kidney diseases. Cluster members are characterized by a high level of creatinine, increased osmolarity, and low hemoglobin and hematocrit values. While such traits are found in patients with chronic diseases, they are highly suggestive of patients with chronic kidney disease, in particular [8], [9]. Cluster 3 is enriched with smokers, and is characterized by individuals with high hemoglobin, red blood cell and hematorcrit levels, a pattern characteristic of sufferers of lung diseases (e.g., smokers and individuals exposed to severely polluted air [10]). We note that for other clusters, correspondence may be missed if it involves health/lifestyle traits that were not recorded in the NHANES dataset, or if careful inspection by an expert in the appropriate medical field is required to note the correspondence.


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

Markovich Gordon M, Moser AM, Rubin E - PLoS ONE (2012)

Selected clusters from the NHANES training set.(A) The mean and standard deviation of biomarker values are shown for three selected clusters generated with 4152 males 20 years of age or older from the NHANES training set, using the Z-score normalized/CLICK pre-processing/clustering combination. For each cluster, the total number of individuals (top, right) and selected health/lifestyle traits that are significantly enriched in that cluster (top, left) are provided. For each enriched term, the enrichment factor (i.e. the frequency of the term in a cluster divided by its frequency in the entire dataset) is also provided. (B) Comparison of the original values of selected biomarkers in clusters 1, 3 and 14. The values for Hb are enlarged in the top middle section of the figure.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3293863&req=5

pone-0029578-g002: Selected clusters from the NHANES training set.(A) The mean and standard deviation of biomarker values are shown for three selected clusters generated with 4152 males 20 years of age or older from the NHANES training set, using the Z-score normalized/CLICK pre-processing/clustering combination. For each cluster, the total number of individuals (top, right) and selected health/lifestyle traits that are significantly enriched in that cluster (top, left) are provided. For each enriched term, the enrichment factor (i.e. the frequency of the term in a cluster divided by its frequency in the entire dataset) is also provided. (B) Comparison of the original values of selected biomarkers in clusters 1, 3 and 14. The values for Hb are enlarged in the top middle section of the figure.
Mentions: To test whether the observed clusters can define biomedically homogenous sub-populations, we inspected other cluster characteristics. We hypothesized that if individuals in the same cluster share similar properties, then the enrichment we observed may correspond to a specific pattern of biomarker values in that cluster. To demonstrate that such correspondence can be observed, the mean biomarker values of three clusters were analyzed in the context of their health/lifestyle traits enrichment (Figure 2). Cluster 14 is highly enriched with diabetic individuals. As expected, members of this cluster are characterized by high levels of glucose, blood osmolarity and triglycerides, traits that are indeed hallmarks of unbalanced diabetic patients [7]. Cluster 1 is enriched with individuals suffering from kidney diseases. Cluster members are characterized by a high level of creatinine, increased osmolarity, and low hemoglobin and hematocrit values. While such traits are found in patients with chronic diseases, they are highly suggestive of patients with chronic kidney disease, in particular [8], [9]. Cluster 3 is enriched with smokers, and is characterized by individuals with high hemoglobin, red blood cell and hematorcrit levels, a pattern characteristic of sufferers of lung diseases (e.g., smokers and individuals exposed to severely polluted air [10]). We note that for other clusters, correspondence may be missed if it involves health/lifestyle traits that were not recorded in the NHANES dataset, or if careful inspection by an expert in the appropriate medical field is required to note the correspondence.

Bottom Line: It has been proposed that clustering clinical markers, such as blood test results, can be used to stratify patients.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.However, optimization of the pre-processing and clustering process may be still required.

View Article: PubMed Central - PubMed

Affiliation: Sharaga Segal Dept of Microbiology and Immunology, Ben Gurion University of the Negev, Beersheba, Israel.

ABSTRACT

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.

Show MeSH
Related in: MedlinePlus