Distance Assisted Recursive Testing

Li, Xuechan; Sung, Anthony; Xie, Jichun

doi:10.48550/arxiv.2103.11085

Cited by 2 publications

(2 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The framework of IF-PCA only assumes feature sparsity but no other particular structures on the features. It is possible that the features are grouped (Chang et al, 2017) or have some tree structures (Li et al, 2021). How to adapt IF-PCA to this setting is an interesting yet open research direction.…”

Section: Discussionmentioning

confidence: 99%

Subject clustering by IF-PCA and several recent methods

Chen¹,

Jin

Ke³

2023

Front. Genet.

View full text Add to dashboard Cite

Subject clustering (i.e., the use of measured features to cluster subjects, such as patients or cells, into multiple groups) is a problem of significant interest. In recent years, many approaches have been proposed, among which unsupervised deep learning (UDL) has received much attention. Two interesting questions are 1) how to combine the strengths of UDL and other approaches and 2) how these approaches compare to each other. We combine the variational auto-encoder (VAE), a popular UDL approach, with the recent idea of influential feature-principal component analysis (IF-PCA) and propose IF-VAE as a new method for subject clustering. We study IF-VAE and compare it with several other methods (including IF-PCA, VAE, Seurat, and SC3) on 10 gene microarray data sets and eight single-cell RNA-seq data sets. We find that IF-VAE shows significant improvement over VAE, but still underperforms compared to IF-PCA. We also find that IF-PCA is quite competitive, slightly outperforming Seurat and SC3 over the eight single-cell data sets. IF-PCA is conceptually simple and permits delicate analysis. We demonstrate that IF-PCA is capable of achieving phase transition in a rare/weak model. Comparatively, Seurat and SC3 are more complex and theoretically difficult to analyze (for these reasons, their optimality remains unclear).

show abstract

Section: Discussionmentioning

confidence: 99%

Subject clustering by IF-PCA and several recent methods

Chen¹,

Jin

Ke³

2023

Front. Genet.

View full text Add to dashboard Cite

show abstract

“…Third, we advise adhering to the tuning parameter selection criteria described in Supporting Information Note 2 for determining the layer number L and leaf size M. These parameters have been optimized based on our prior studies on hierarchical multiple testing (Li et al, 2023;Pura et al, 2023). Although increasing layer numbers could potentially lead to more discoveries, it may also elevate the FDR.…”

Section: Computation Timementioning

confidence: 99%

DYNATE: Localizing rare‐variant association regions via multiple testing embedded in an aggregation tree

Li,

Pura,

Allen

et al. 2023

Genetic Epidemiology

View full text Add to dashboard Cite

Rare‐variants (RVs) genetic association studies enable researchers to uncover the variation in phenotypic traits left unexplained by common variation. Traditional single‐variant analysis lacks power; thus, researchers have developed various methods to aggregate the effects of RVs across genomic regions to study their collective impact. Some existing methods utilize a static delineation of genomic regions, often resulting in suboptimal effect aggregation, as neutral subregions within the test region will result in an attenuation of signal. Other methods use varying windows to search for signals but often result in long regions containing many neutral RVs. To pinpoint short genomic regions enriched for disease‐associated RVs, we developed a novel method, DYNamic Aggregation TEsting (DYNATE). DYNATE dynamically and hierarchically aggregates smaller genomic regions into larger ones and performs multiple testing for disease associations with a controlled weighted false discovery rate. DYNATE's main advantage lies in its strong ability to identify short genomic regions highly enriched for disease‐associated RVs. Extensive numerical simulations demonstrate the superior performance of DYNATE under various scenarios compared with existing methods. We applied DYNATE to an amyotrophic lateral sclerosis study and identified a new gene, EPG5, harboring possibly pathogenic mutations.

show abstract

Distance Assisted Recursive Testing

Cited by 2 publications

References 24 publications

Subject clustering by IF-PCA and several recent methods

Subject clustering by IF-PCA and several recent methods

DYNATE: Localizing rare‐variant association regions via multiple testing embedded in an aggregation tree

Contact Info

Product

Resources

About