Sparse partial least squares with group and subgroup structure.

Stat Med

ARC Centre of Excellence for Mathematical and Statistical Frontiers, Queensland University of Technology, Brisbane, Australia.

Published: October 2018


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

Integrative analysis of high dimensional omics datasets has been studied by many authors in recent years. By incorporating prior known relationships among the variables, these analyses have been successful in elucidating the relationships between different sets of omics data. In this article, our goal is to identify important relationships between genomic expression and cytokine data from a human immunodeficiency virus vaccine trial. We proposed a flexible partial least squares technique, which incorporates group and subgroup structure in the modelling process. Our new method accounts for both grouping of genetic markers (eg, gene sets) and temporal effects. The method generalises existing sparse modelling techniques in the partial least squares methodology and establishes theoretical connections to variable selection methods for supervised and unsupervised problems. Simulation studies are performed to investigate the performance of our methods over alternative sparse approaches. Our R package sgspls is available at https://github.com/matt-sutton/sgspls.

Download full-text PDF

Source
http://dx.doi.org/10.1002/sim.7821DOI Listing

Publication Analysis

Top Keywords

partial squares
12
group subgroup
8
subgroup structure
8
sparse partial
4
squares group
4
structure integrative
4
integrative analysis
4
analysis high
4
high dimensional
4
dimensional omics
4

Similar Publications

River water quality degradation is a prevailing problem in coastal China with intensifying human-nature interaction. However, the spatial and temporal dynamics of water quality and their drivers remain poorly understood. In this study, we developed an analytical framework integrating self-organizing mapping (SOM) with partial least squares structural equation models (PLS-SEMs) to analyze the patterns and drivers of river water quality at 49 stations from 2021 to 2023 in Fujian Province, a coastal region in southeastern China.

View Article and Find Full Text PDF

Assessment of yerba mate quality based on branch content via digital image analysis.

Food Chem

September 2025

Group of Chemical Analysis and Chemometrics, Department of Chemistry, Federal University of Paraná, P.O. Box: 19032, Curitiba, PR 81531-980, Brazil. Electronic address:

Yerba mate, a key crop in South America, is prized for its pleasant taste and high organoleptic quality, often linked to lower branch content. To quantify branch content and authenticate high-quality samples (less than 30 % m/m branch content), a Chemometrics-assisted Color Histogram-based Analytical System (CACHAS) was employed. Using Hue-Saturation-Value (HSV) histograms, Partial Least Squares (PLS) demonstrated excellent predictive performance, achieving a root mean square error (RMSEP) of 4.

View Article and Find Full Text PDF

Organophosphorus nerve agents (OPNAs), including G-agents, EGA (ethyltabun, phosphonamidic acid, P-cyano-N,N-diethyl-, ethyl ester) and V-agents, VM (O-ethyl S-(2-diethylaminoethyl) phosphonothiolate), are highly toxic chemical warfare agents (CWAs) with severe risks to human health and environmental security. This study proposes a chemometric-driven framework for forensic tracing of their synthetic pathways using high-resolution GC × GC-TOFMS. By integrating advanced statistical analysis, we identified 160 synthesis-associated chemical attribution signatures (CAS) for EGA and 138 process-specific CAS for VM, with 11 overlapping markers, including ethoxyphosphates and diethylaminoethylamine derivatives.

View Article and Find Full Text PDF

Brain activation for language and its relationship to cognitive and linguistic measures.

Cereb Cortex

August 2025

Faculty of Psychology and Education Science, Department of Psychology, University of Geneva, Chemin des Mines 9, Geneva, 1202, Switzerland.

Language learning and use relies on domain-specific, domain-general cognitive and sensory-motor functions. Using fMRI during story listening and behavioral tests, we investigated brain-behavior associations between linguistic and non-linguistic measures in individuals with varied multilingual experience and reading skills, including typical reading participants (TRs) and dyslexic readers (DRs). Partial Least Square Correlation revealed a main component linking cognitive, linguistic, and phonological measures to amodal/associative brain areas.

View Article and Find Full Text PDF

Purpose: Crohn's disease (CD) is characterized by enteric inflammation, often resulting in strictures and penetrating complications, which may alter patient management prior to the initiation of biologic therapy. Our aim is to assess the frequency of missed stricturing and internal penetrating complications in CD patients on computed tomography enterography (CTE) and magnetic resonance enterography (MRE) performed prior to anti-TNF therapy.

Methods: We retrospectively reviewed patients from two tertiary centers who underwent CTE\MRE within six months before starting anti-TNF therapy.

View Article and Find Full Text PDF