A Framework for Multi-Objective Clustering and Its Application to Co-Location Mining

Jiamthapthaksin, Rachsuda; Eick, Christoph F.; Vilalta, Ricardo

doi:10.1007/978-3-642-03348-3_20

Cited by 11 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-objective clustering (MOC) decomposes a data set into related groups, maximizing multiple objectives in parallel. Several frameworks exist to implement MOC, MobiLLE relies on multi-run clustering, where a clustering algorithm runs multiple times to optimize different objectives that capture a compound fitness function ( 13 ). MobiLLE proceeds through two main steps: pre-clustering and refinement.…”

Section: Methodsmentioning

confidence: 99%

A multi-objective based clustering for inferring BCR clones from high-throughput B cell repertoire data

Abdollahi

Septenville

Ripoche

et al. 2021

Preprint

View full text Add to dashboard Cite

The adaptive B cell response is driven by the expansion, somatic hypermutation, and selection of B cell clones. A high number of clones in a B cell population indicates a highly diverse repertoire, while clonal size distribution and sequence diversity within clones can be related to antigen's selective pressure. Identifying clones is fundamental to many repertoire studies, including repertoire comparisons, clonal tracking and statistical analysis. Several methods have been developed to group sequences from high-throughput B cell repertoire data. Current methods use clustering algorithms to group clonally-related sequences based on their similarities or distances. Such approaches create groups by optimizing a single objective that typically minimizes intra-clonal distances. However, optimizing several objective functions can be advantageous and boost the algorithm convergence rate. Here we propose a new method based on multi-objective clustering. Our approach requires V(D)J annotations to obtain the initial clones and iteratively applies two objective functions that optimize cohesion and separation within clones simultaneously. We show that under simulations with varied mutation rates, our method greatly improves clonal grouping as compared to other tools. When applied to experimental repertoires generated from high-throughput sequencing, its clustering results are comparable to the most performing tools. The method based on multi-objective clustering can accurately identify clone members, has fewer parameter settings and presents the lowest running time among existing tools. All these features constitute an attractive option for repertoire analysis, particularly in the clinical context to unravel the mechanisms involved in the development and evolution of B cell malignancies.

show abstract

Section: Methodsmentioning

confidence: 99%

A multi-objective based clustering for inferring BCR clones from high-throughput B cell repertoire data

Abdollahi

Septenville

Ripoche

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Multi-objective clustering (MOC) decomposes a dataset into related groups, maximizing multiple objectives in parallel. Several frameworks exist to implement MOC, MobiLLe relies on multi-run clustering, where a clustering algorithm runs multiple times to optimize different objectives that capture a compound fitness function [19]. MobiLLe proceeds through two main steps: pre-clustering and refinement.…”

Section: Mobillementioning

confidence: 99%

A multi-objective based clustering for inferring BCR clonal lineages from high-throughput B cell repertoire data

et al. 2022

View full text Add to dashboard Cite

The adaptive B cell response is driven by the expansion, somatic hypermutation, and selection of B cell clonal lineages. A high number of clonal lineages in a B cell population indicates a highly diverse repertoire, while clonal size distribution and sequence diversity antigen selective pressure. Identifying clonal lineages is fundamental to many repertoire studies, including repertoire comparisons, clonal tracking, and statistical analysis. Several methods have been developed to group sequences from high-throughput B cell repertoire data. Current methods use clustering algorithms to group clonally-related sequences based on their similarities or distances. Such approaches create groups by optimizing a single objective that typically minimizes intra-clonal distances. However, optimizing several objective functions can be advantageous and boost the algorithm convergence rate. Here we propose a new method based on multi-objective clustering. Our approach requires V(D)J annotations to obtain the initial groups and iteratively applies two objective functions that optimize cohesion and separation within clonal lineages simultaneously. We show that our method greatly improves clonal lineage grouping on simulated benchmarks with varied mutation rates compared to other tools. When applied to experimental repertoires generated from high-throughput sequencing, its clustering results are comparable to the most performing tools and can reproduce the results of previous publications. The method based on multi-objective clustering can accurately identify clonally-related antibody sequences and presents the lowest running time among state-of-art tools. All these features constitute an attractive option for repertoire analysis, particularly in the clinical context. MobiLLe can potentially help unravel the mechanisms involved in developing and evolving B cell malignancies.

show abstract

“…There is significant research, of late, on multiobjective clustering. Jiamthapthaksinet al [1] focused on colocation mining with the help of MOC. They proposed a framework towards it in order to identity regions in which co-located Arsenic concentrations exist.…”

Section: Related Workmentioning

confidence: 99%

“…Multi-objective clustering (MOC) is aimed at dividing given data instances into similar groups while meeting multiple objective functions in parallel. MOC improves capabilities of clustering with different objectives, fitness functions and thresholds [1]. MOC promotes spatial co-location mining besides supporting various real world applications.…”

Section: Introductionmentioning

confidence: 99%

MOCF: A Multi-Objective Clustering Framework using an Improved Particle Swarm Optimization Algorithm

Yedukondalu

Samunnisa²,

Bhavsingh³

et al. 2022

IJRITCC

View full text Add to dashboard Cite

Traditional clustering algorithms, such as K-Means, perform clustering with a single goal in mind. However, in many real-world applications, multiple objective functions must be considered at the same time. Furthermore, traditional clustering algorithms have drawbacks such as centroid selection, local optimal, and convergence. Particle Swarm Optimization (PSO)-based clustering approaches were developed to address these shortcomings. Animals and their social Behaviour, particularly bird flocking and fish schooling, inspire PSO. This paper proposes the Multi-Objective Clustering Framework (MOCF), an improved PSO-based framework. As an algorithm, a Particle Swarm Optimization (PSO) based Multi-Objective Clustering (PSO-MOC) is proposed. It significantly improves clustering efficiency. The proposed framework's performance is evaluated using a variety of real-world datasets. To test the performance of the proposed algorithm, a prototype application was built using the Python data science platform. The empirical results showed that multi-objective clustering outperformed its single-objective counterparts.

show abstract

A Framework for Multi-Objective Clustering and Its Application to Co-Location Mining

Cited by 11 publications

References 12 publications

A multi-objective based clustering for inferring BCR clones from high-throughput B cell repertoire data

A multi-objective based clustering for inferring BCR clones from high-throughput B cell repertoire data

A multi-objective based clustering for inferring BCR clonal lineages from high-throughput B cell repertoire data

MOCF: A Multi-Objective Clustering Framework using an Improved Particle Swarm Optimization Algorithm

Contact Info

Product

Resources

About