Gene expression clustering using local neighborhood-based similarity measures

Jothi, R.; Mohanty, Sraban Kumar; Ojha, Aparajita

doi:10.1016/j.compeleceng.2021.107032

Cited by 8 publications

(9 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A study has shown the three most important GO terms can indicate the biological relevance of clustering [16]. To further understand the biological significance of the clustered gene modules generated by the proposed method, we performed GO analysis on the categories obtained from K-means clustering based on the DINNSM method.…”

Section: Biological Significancementioning

confidence: 99%

“…Hierarchical clustering is effective for retrieving the nested relationships of gene expression data [6,16]. It creates a nested hierarchical clustering tree by calculating the similarity between different categories of data points.…”

Section: Clustering Algorithmsmentioning

confidence: 99%

“…FCM is the most widely used fuzzy clustering algorithm based on objective function [16]. It first calculates the membership matrix of samples (genes).…”

Section: Clustering Algorithmsmentioning

confidence: 99%

“…Clustering algorithms are the most widely used methods for detecting modules in gene expression data. They can roughly be divided into partitioning, hierarchical, density-based, model-based, and graph-based methods [8,16,17], as well as other clustering methods such as bicluster andδ-clusters [18][19][20]. Although the clustering analysis has been successful in many aspects, there are still some restrictions in gene expression data analysis due to the high dimensionality and noise of the data.…”

mentioning

confidence: 99%

“…This method preserves the similarity information of the expression level's trajectory pattern over time intervals. At the same time, it considers the similarity information of the trajectory patterns at the time points where both profiles reach their maximum and minimum expression levels, respectively; Jothi et al [16] proposed an approximate similarity measurement method based on minimum spanning tree local neighborhood representation; Hou et al [32] applied Distance correlation (DC) to weighted gene co-expression network analysis. This method enhances enrichment analysis results, improving module stability, but it is time-consuming and has no validation with the evaluation methods.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Clustering Algorithm Based on Dual-Index Nearest Neighbor Similarity Measure and Its Application in Gene Expression Data Analysis

song

Yang

et al. 2023

Preprint

View full text Add to dashboard Cite

Background The critical step in analyzing gene expression data is to divide genes into co-expression modules using module detection methods. Clustering algorithms are the most commonly employed technique for gene module detection. To obtain gene modules with great biological significance, the choice of an appropriate similarity measure methodology is vital. However, commonly used similarity measurement may not fully capture the complexities of biological systems. Hence, exploring more informative similarity measures before partitioning gene co-expression modules remains important. Results In this paper, we proposed a Dual-Index Nearest Neighbor Similarity Measure (DINNSM) algorithm to address the above issue. The algorithm first calculates the similarity matrix between genes using Pearson correlation or Spearman correlation. Then, nearest neighbor measurements are constructed based on the similarity matrix. Finally, the similarity matrix is reconstructed. We tested the six similarity measurement methods (Pearson correlation, Spearman correlation, Euclidean distance, maximum information coefficient, distance correlation, and DINNSM) by using four clustering algorithms: K-means, Hierarchical, FCM, and WGCNA on three independent gene expression datasets. The cluster evaluation was based on four indices: the Silhouette index, Calinski-Harabaz index, Adjust-Biological homogeneity index, and Davies-Bouldin index. The results showed that DINNSM is accurate and can get biologically meaningful gene co-expression modules. Conclusions DINNSM is better at revealing the complex biological relationships between genes and helps to obtain more accurate and biologically meaningful gene co-expression modules.

show abstract

Section: Biological Significancementioning

confidence: 99%

Section: Clustering Algorithmsmentioning

confidence: 99%

“…FCM is the most widely used fuzzy clustering algorithm based on objective function [16]. It first calculates the membership matrix of samples (genes).…”

Section: Clustering Algorithmsmentioning

confidence: 99%