Greedy feature selection for glycan chromatography data with the generalized Dirichlet distribution

Abstract

BackgroundGlycoproteins are involved in a diverse range of biochemical and biological processes. Changes in protein glycosylation are believed to occur in many diseases, particularly during cancer initiation and progression. The identification of biomarkers for human disease states is becoming increasingly important, as early detection is key to improving survival and recovery rates. To this end, the serum glycome has been proposed as a potential source of biomarkers for different types of cancers.High-throughput hydrophilic interaction liquid chromatography (HILIC) technology for glycan analysis allows for the detailed quantification of the glycan content in human serum. However, the experimental data from this analysis is compositional by nature. Compositional data are subject to a constant-sum constraint, which restricts the sample space to a simplex. Statistical analysis of glycan chromatography datasets should account for their unusual mathematical properties.As the volume of glycan HILIC data being produced increases, there is a considerable need for a framework to support appropriate statistical analysis. Proposed here is a methodology for feature selection in compositional data. The principal objective is to provide a template for the analysis of glycan chromatography data that may be used to identify potential glycan biomarkers.ResultsA greedy search algorithm, based on the generalized Dirichlet distribution, is carried out over the feature space to search for the set of “grouping variables” that best discriminate between known group structures in the data, modelling the compositional variables using beta distributions. The algorithm is applied to two glycan chromatography datasets. Statistical classification methods are used to test the ability of the selected features to differentiate between known groups in the data. Two well-known methods are used for comparison: correlation-based feature selection (CFS) and recursive partitioning (rpart). CFS is a feature selection method, while recursive partitioning is a learning tree algorithm that has been used for feature selection in the past.ConclusionsThe proposed feature selection method performs well for both glycan chromatography datasets. It is computationally slower, but results in a lower misclassification rate and a higher sensitivity rate than both correlation-based feature selection and the classification tree method.

Choosing proper normalization is essential for discovery of sparse glycan biomarkers

“…It is often claimed that glycans are by their nature compositions, and that percentage of glycan species in the whole is biologically relevant information. 28 To dispute such claims is not of our interest; our intention is to increase the awareness of spurious correlations caused by row-wise normalization. In this work, we did not try to disentangle the true correlation structure in view of biochemical pathways, but rather demonstrated that choosing one normalization method can cause several potential issues and problems for downstream analysis.…”

Factors influencing job preferences of health workers providing obstetric care: results from discrete choice experiments in Malawi, Mozambique and Tanzania

“…A similar variable selection strategy to the above was used in Raftery and Dean [34] and in Galligan et al [10] to select variables for inclusion in clustering and classification models respectively. Results across the 200 bootstrap samples were compiled.…”

Association Between Human Gut Microbiome and N-Glycan Composition of Total Plasma Proteome

“…These derived traits average glycosylation features such as branching, galactosylation, and sialylation across different individual glycan structures, and consequently, they may be more closely related to individual enzymatic activity. For the original traits, CLR transformation from the "compositions" R package (van den Boogaart and Tolosana-Delgado, 2008) was implemented to account for the compositional nature of the data (Galligan et al, 2013). For the derived traits, different approaches of compositional transformations were used depending on the type of the features (Supplementary Table 6).…”