Data normalization and scaling: Consequences for the analysis in omics sciences

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989592%3A15310%2F18%3A73589128" target="_blank" >RIV/61989592:15310/18:73589128 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Data normalization and scaling: Consequences for the analysis in omics sciences

Popis výsledku v původním jazyce

The main task in the analysis of omics data is to understand biological information in the data. From a statistical point of view, classification analysis is one of the goals. If the data are consisting of groups of, e.g., controls and patients, the accurate prediction of new samples is desirable. To understand the processes in the human body or in other organisms, the interpretation of the model is necessary. In the two-group setting, the information about important features is one of the main tasks in omics disciplines. In metabolomics, the problem is called biomarker identification, while in genetics this is called fold changes problem, where it is examined for a feature, how many times the average concentration in one group is higher/lower than for the other group. In statistics, this is often referred to as the feature selection problem. Section 4 analyzes the impact of pretreatment methods on publicly available real-world data sets in terms of classification and feature selection analysis. As an example of omics disciplines, the data sets are originating from the metabolomics field. Section 5 discusses and summarizes the main findings and provides some overall recommendations.

Název v anglickém jazyce

Data normalization and scaling: Consequences for the analysis in omics sciences

Popis výsledku anglicky

The main task in the analysis of omics data is to understand biological information in the data. From a statistical point of view, classification analysis is one of the goals. If the data are consisting of groups of, e.g., controls and patients, the accurate prediction of new samples is desirable. To understand the processes in the human body or in other organisms, the interpretation of the model is necessary. In the two-group setting, the information about important features is one of the main tasks in omics disciplines. In metabolomics, the problem is called biomarker identification, while in genetics this is called fold changes problem, where it is examined for a feature, how many times the average concentration in one group is higher/lower than for the other group. In statistics, this is often referred to as the feature selection problem. Section 4 analyzes the impact of pretreatment methods on publicly available real-world data sets in terms of classification and feature selection analysis. As an example of omics disciplines, the data sets are originating from the metabolomics field. Section 5 discusses and summarizes the main findings and provides some overall recommendations.

Druh

C - Kapitola v odborné knize

CEP obor

—

OECD FORD obor

10103 - Statistics and probability

Projekt

<a href="/cs/project/GF15-34613L" target="_blank" >GF15-34613L: Statistika v metabolomice pro výzkum biomarkerů v medicíně</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název knihy nebo sborníku

Data analysis for omics sciences: Methods and applications

ISBN

978-0-444-64044-4

Počet stran výsledku

32

Strana od-do

165-196

Počet stran knihy

706

Název nakladatele

Elsevier

Místo vydání

Amsterdam

Kód UT WoS kapitoly

—