2015
DOI: 10.1142/s0219720015500183
|View full text |Cite
|
Sign up to set email alerts
|

Abstract: Transcript-level quantification is often measured across two groups of patients to aid the discovery of biomarkers and detection of biological mechanisms involving these biomarkers. Statistical tests lack power and false discovery rate is high when sample size is small. Yet, many experiments have very few samples (≤ 5). This creates the impetus for a method to discover biomarkers and mechanisms under very small sample sizes. We present a powerful method, ESSNet, that is able to identify subnetworks consistentl… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
25
0

Year Published

2015
2015
2023
2023

Publication Types

Select...
6
1

Relationship

3
4

Authors

Journals

citations
Cited by 30 publications
(25 citation statements)
references
References 39 publications
(47 reference statements)
0
25
0
Order By: Relevance
“…Whilst design is nontrivial and requires proper integration of bio-statistics, networks and proteomics [4,[6][7][8][9], networkbased approaches are already contributing towards resolving idiosyncratic coverage and consistency problems in clinical proteomics [10][11][12]. Soh et al [13] and Lim et al [14,15] have further demonstrated that network-based feature-selection methods are highly reproducible, and select phenotypically relevant features. Additionally, as a special instance of subnetworks, Goh et al demonstrated that known protein complexes are highly enriched for biological signal [16], and outperforms predicted complexes from reference networks e.g.…”
Section: Significance Of the Studymentioning
confidence: 99%
See 1 more Smart Citation
“…Whilst design is nontrivial and requires proper integration of bio-statistics, networks and proteomics [4,[6][7][8][9], networkbased approaches are already contributing towards resolving idiosyncratic coverage and consistency problems in clinical proteomics [10][11][12]. Soh et al [13] and Lim et al [14,15] have further demonstrated that network-based feature-selection methods are highly reproducible, and select phenotypically relevant features. Additionally, as a special instance of subnetworks, Goh et al demonstrated that known protein complexes are highly enriched for biological signal [16], and outperforms predicted complexes from reference networks e.g.…”
Section: Significance Of the Studymentioning
confidence: 99%
“… and Lim et al. have further demonstrated that network‐based feature‐selection methods are highly reproducible, and select phenotypically relevant features. Additionally, as a special instance of subnetworks, Goh et al.…”
Section: Introductionmentioning
confidence: 99%
“…Local analysis is more easily steered towards functional applications. In particular, if only a small part of a network is phenotypically relevant, then global analysis may result in loss of signal [18,19]. Moreover, it is easier to functionally characterize small and manageable local sub-networks, and to use these as predictive features.…”
Section: Forget Whole Network Focus On Functional Unitsmentioning
confidence: 99%
“…Feature selection at the level of networks can resolve this stability issue. Lim et al demonstrated across a large variety of genomics datasets that, despite numerous resampling and multiple sample batches, the differential networks always remain stable [18,43]. The networks act as a buffer against random expression fluctuations at the individual protein level.…”
Section: Box 2 Feature-selection Reproducibility and Class-predictiomentioning
confidence: 99%
“…Rank-based methods are powerful: Rank-based approaches have been shown to be more robust than those utilizing full expressional information. Recent evaluations by Patil et al (Patil et al 2015) and Lim et al (Lim et al 2015) demonstrated that the features selected by rank-based approaches are stable and generalizable onto other similar datasets. Building upon this, RBNAs, which utilize rank information, have been shown to be extremely powerful for identification of relevant features, producing unparallelled prediction reliability and reproducibility in transcriptomics studies (Soh et al 2011;Lim and Wong 2014).…”
Section: Introductionmentioning
confidence: 99%