High-Dimensional Multi-Block Analysis of Factors Associated with Thrombin Generation Potential

Lorenzo, Hadrien; Razzaq, Misbah; Odeber, Jacob; Morange, Pierre‐Emmanuel; Saracco, Jérôme; Trégouët, David‐Alexandre; Thiébaut, Rodolphe

doi:10.1109/cbms.2019.00094

Cited by 3 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi‐Block Data‐Driven sparse partial least squares (mdd‐sPLS) (Lorenzo, Razzaq, et al, 2019; Lorenzo, Saracco, & Thiébaut, 2019) is a complex method based on the partial least squares (PLS) technique. Both the training data and the test data are imputed, where these imputations are performed separately.…”

Section: Existing Prediction Approaches For Partly Missing Multi‐omic...mentioning

confidence: 99%

“…Thung et al (2014) do allow missing values in the test data, but the missing data in the training and test data are imputed jointly, which is why retraining is also required here. For mdd‐sPLS (Lorenzo, Razzaq, et al, 2019; Lorenzo, Saracco, & Thiébaut, 2019), separate imputation is performed on both training and test data. However, since the mdd‐PLS procedure involves predicting the outcome, it needs to be reapplied as well when new test data are received.…”

Section: Existing Prediction Approaches For Partly Missing Multi‐omic...mentioning

confidence: 99%

“…Moreover, we used one component for the involved matrix decomposition and inversely weighted the components per block by the number of selected covariates per block. See Lorenzo, Saracco, and Thiébaut (2019) for details.…”

Section: Empirical Comparison Study Of Approaches For Partly Missing ...mentioning

confidence: 99%

See 2 more Smart Citations

Prediction approaches for partly missing multi‐omics covariate data: A literature review and an empirical comparison study

Hornung

Ludwigs

Hagenberg

et al. 2023

WIREs Computational Stats

View full text Add to dashboard Cite

As the availability of omics data has increased in the last few years, more multi‐omics data have been generated, that is, high‐dimensional molecular data consisting of several types such as genomic, transcriptomic, or proteomic data, all obtained from the same patients. Such data lend themselves to being used as covariates in automatic outcome prediction because each omics type may contribute unique information, possibly improving predictions compared to using only one omics data type. Frequently, however, in the training data and the data to which automatic prediction rules should be applied, the test data, the different omics data types are not available for all patients. We refer to this type of data as block‐wise missing multi‐omics data. First, we provide a literature review on existing prediction methods applicable to such data. Subsequently, using a collection of 13 publicly available multi‐omics data sets, we compare the predictive performances of several of these approaches for different block‐wise missingness patterns. Finally, we discuss the results of this empirical comparison study and draw some tentative conclusions.This article is categorized under: Applications of Computational Statistics > Genomics/Proteomics/Genetics Applications of Computational Statistics > Health and Medical Data/Informatics Statistical and Graphical Methods of Data Analysis > Analysis of High Dimensional Data

show abstract

Section: Existing Prediction Approaches For Partly Missing Multi‐omic...mentioning

confidence: 99%

Section: Existing Prediction Approaches For Partly Missing Multi‐omic...mentioning

confidence: 99%

See 1 more Smart Citation

Prediction approaches for partly missing multi‐omics covariate data: A literature review and an empirical comparison study

Hornung

Ludwigs

Hagenberg

et al. 2023

WIREs Computational Stats

View full text Add to dashboard Cite

show abstract

“…Discriminative performances and adjustment for optimism will be performed on each cohort. Subsequently, different correlation networks [8], unsupervised clustering and machine/deep learning methodologies [9][10][11] will be applied to identify new biological partners to these candidates. These data will also be integrated into knowledge-based biological databases such as KEGG, Reactome, Gene Ontology database and other publicly available "omics" resources.…”

Section: Derivation Of the Optimal Biomarker Signaturementioning

confidence: 99%

Design and Rationale of the ERA-CVD Consortium PREMED-CAD—Precision Medicine in Coronary Artery Disease

et al. 2020

Self Cite

View full text Add to dashboard Cite

Cardiovascular diseases (CVDs) comprise 45% of all deaths in Europe and causes 3.9 million deaths annually. Coronary artery disease (CAD) which includes myocardial infarction (MI) represents the most common form of CVD. A relevant proportion of MI cases seems preventable since reports claim that up to two-thirds of these patients exhibit symptoms suggestive for MI within 12 months prior to the acute MI event. An early identification of these at-risk subjects is necessary to manage an early and efficient treatment during the ischemic phase. The aim of the PRecision MEDicine in Coronary Artery Disease (PREMED-CAD) consortium is to apply a system medicine approach towards studying and identifying an ischemia specific ‘biomarker signature’ that improves the identification of individuals ‘at-risk’ for acute MI. The consortium will take an interdisciplinary and translational approach integrating knowledge from CAD epidemiology, imaging, bioinformatics, statistics and molecular biology, as well as existing phenotypic, blood-based and clinical biomarker data of distinct CAD and subclinical MI phenotypes. This biomarker signature will be validated through atherosclerosis-prone mouse models and human cohorts. The validated signature will be translated in a real-world clinical setting using an ongoing clinical trial comprising patients with subclinical ischemia. The aim of the knowledge obtained from this project is to aid in early MI detection and reduce the mortality and morbidity rate in these at-risk MI individuals.

show abstract

“…Thus, the aim of the present work was to identify, from among a pool of 20 biomarkers of interest measured in five different muscles, whether a combination of biomarkers could predict the tenderness of each of the five muscles. Therefore, a specific statistical methodology adapted to a low number of samples with a large number of observations [6,7] was tested.…”

Section: Introductionmentioning

confidence: 99%

An Original Methodology for the Selection of Biomarkers of Tenderness in Five Different Muscles

Ellies-Oury

Lorenzo

Denoyelle

et al. 2019

Foods

Self Cite

View full text Add to dashboard Cite

For several years, studies conducted for discovering tenderness biomarkers have proposed a list of 20 candidates. The aim of the present work was to develop an innovative methodology to select the most predictive among this list. The relative abundance of the proteins was evaluated on five muscles of 10 Holstein cows: gluteobiceps, semimembranosus, semitendinosus, Triceps brachii and Vastus lateralis. To select the most predictive biomarkers, a multi-block model was used: The Data-Driven Sparse Partial Least Square. Semimembranosus and Vastus lateralis muscles tenderness could be well predicted (R2 = 0.95 and 0.94 respectively) with a total of 7 out of the 5 times 20 biomarkers analyzed. An original result is that the predictive proteins were the same for these two muscles: µ-calpain, m-calpain, h2afx and Hsp40 measured in m. gluteobiceps and µ-calpain, m-calpain and Hsp70-8 measured in m. Triceps brachii. Thus, this method is well adapted to this set of data, making it possible to propose robust candidate biomarkers of tenderness that need to be validated on a larger population.

show abstract

High-Dimensional Multi-Block Analysis of Factors Associated with Thrombin Generation Potential

Cited by 3 publications

References 9 publications

Prediction approaches for partly missing multi‐omics covariate data: A literature review and an empirical comparison study

Prediction approaches for partly missing multi‐omics covariate data: A literature review and an empirical comparison study

Design and Rationale of the ERA-CVD Consortium PREMED-CAD—Precision Medicine in Coronary Artery Disease

An Original Methodology for the Selection of Biomarkers of Tenderness in Five Different Muscles

Contact Info

Product

Resources

About