Focused multidimensional scaling: interactive visualization for exploration of high-dimensional data.
Clustering
High-dimensional data
Personalized medicine
Visualization
Journal
BMC bioinformatics
ISSN: 1471-2105
Titre abrégé: BMC Bioinformatics
Pays: England
ID NLM: 100965194
Informations de publication
Date de publication:
02 May 2019
02 May 2019
Historique:
received:
11
12
2018
accepted:
27
03
2019
entrez:
4
5
2019
pubmed:
3
5
2019
medline:
14
6
2019
Statut:
epublish
Résumé
Visualization is an important tool for generating meaning from scientific data, but the visualization of structures in high-dimensional data (such as from high-throughput assays) presents unique challenges. Dimension reduction methods are key in solving this challenge, but these methods can be misleading- especially when apparent clustering in the dimension-reducing representation is used as the basis for reasoning about relationships within the data. We present two interactive visualization tools, distnet and focusedMDS, that help in assessing the validity of a dimension-reducing plot and in interactively exploring relationships between objects in the data. The distnet tool is used to examine discrepancies between the placement of points in a two dimensional visualization and the points' actual similarities in feature space. The focusedMDS tool is an intuitive, interactive multidimensional scaling tool that is useful for exploring the relationships of one particular data point to the others, that might be useful in a personalized medicine framework. We introduce here two freely available tools for visually exploring and verifying the validity of dimension-reducing visualizations and biological information gained from these. The use of such tools can confirm that conclusions drawn from dimension-reducing visualizations are not simply artifacts of the visualization method, but are real biological insights.
Sections du résumé
BACKGROUND
BACKGROUND
Visualization is an important tool for generating meaning from scientific data, but the visualization of structures in high-dimensional data (such as from high-throughput assays) presents unique challenges. Dimension reduction methods are key in solving this challenge, but these methods can be misleading- especially when apparent clustering in the dimension-reducing representation is used as the basis for reasoning about relationships within the data.
RESULTS
RESULTS
We present two interactive visualization tools, distnet and focusedMDS, that help in assessing the validity of a dimension-reducing plot and in interactively exploring relationships between objects in the data. The distnet tool is used to examine discrepancies between the placement of points in a two dimensional visualization and the points' actual similarities in feature space. The focusedMDS tool is an intuitive, interactive multidimensional scaling tool that is useful for exploring the relationships of one particular data point to the others, that might be useful in a personalized medicine framework.
CONCLUSIONS
CONCLUSIONS
We introduce here two freely available tools for visually exploring and verifying the validity of dimension-reducing visualizations and biological information gained from these. The use of such tools can confirm that conclusions drawn from dimension-reducing visualizations are not simply artifacts of the visualization method, but are real biological insights.
Identifiants
pubmed: 31046657
doi: 10.1186/s12859-019-2780-y
pii: 10.1186/s12859-019-2780-y
pmc: PMC6498510
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
221Subventions
Organisme : NIGMS NIH HHS
ID : R25 GM086262
Pays : United States
Organisme : Deutsche Forschungsgemeinschaft
ID : SFB 1036
Références
Bioinformatics. 2001 Apr;17(4):309-18
pubmed: 11301299
Proc Natl Acad Sci U S A. 2001 Jul 31;98(16):8961-5
pubmed: 11470909
Funct Integr Genomics. 2000 Nov;1(3):156-73
pubmed: 11793234
Bioinformatics. 2002 Nov;18(11):1462-9
pubmed: 12424117
Bioinformatics. 2005 May 15;21(10):2546-7
pubmed: 15713738
Bioinformatics. 2006 Jun 15;22(12):1540-2
pubmed: 16595560
IEEE Trans Vis Comput Graph. 2011 Dec;17(12):2301-9
pubmed: 22034350
Science. 2015 Mar 6;347(6226):1138-42
pubmed: 25700174
Oncotarget. 2017 May 5;8(34):56338-56350
pubmed: 28915594
Nat Biotechnol. 2018 Jun;36(5):411-420
pubmed: 29608179