Practical I – Biomarkers: from single markers to omics
QUESTIONS
1. What is the research question of the study by Chen et al, 2015? Does this study focus on
prediction or on understanding? Explain.
The aim of the study is to identify important metabolites to distinguish a normal metabolic state
from an abnormal obesity metabolome.
Explore the differences between the MHO and MAO groups
The focus of the study is understanding, because they want to find metabolomes that identify
potential metabolic pathways that may regulate the different metabolic characteristics of obesity.
Etiologic research
o Understand mechanisms
o Aim to establish causality
But also prediction
2. Does this study examine a biomarker of exposure, susceptibility, or (intermediate) endpoint?
Explain.
The biomarker is an endpoint, because the biomarker reflect the type of obesity, namely
metabolically healthy obesity (MHO) or metabolic abnormal obesity (MAO) group. So with the use of
metabolites one distinguish MHO and MAO.
Also an exposure and susceptibility marker
3. Are the metabolomics measurements performed targeted or untargeted? Explain the difference.
What are the advantages and disadvantages of both approaches?
Both a targeted and untargeted analysis was employed in this study.
Target measurement = used to estimate the importance of each metabolite within a specific
pathway.
- Hypothesis driven
- The metabolites you want to investigate are already identified (specific)
- Aim = absolute quantification of specific metabolites in biological samples
- Limited number of metabolites
Advantages:
- Used when you want to investigate a specific metabolite
Disadvantages:
- Require a priori knowledge of metabolites of interest
- Do not achieve global coverage
Untargeted measurement = quantify the global metabolic profile of a sample
- Hypothesis-generation
, Practical I – Biomarkers: from single markers to omics
- Collect and analyse all detectable metabolites in a sample
- You do not know the identification of the metabolites
- Compare differences (peaks) in metabolic profiles across different (sub)populations
- As many metabolites as possible
- Relatively easy automated high-throughput
Advantages:
- Reveal novel compound and chemical unknowns
- Leading to new insights in disease mechanisms or treatment
Disadvantages:
- You do not know the identity of the metabolites
Library = database that contains information to identify metabolites
4. Was the sample collection, handling, storage and analysis suitable for measuring
(metabol)omics data? Why (not)? Which issues should be taken into account?
What should be mentioned in the method section:
- Sample collection = blood samples after 8 hours of fasting
o Not mentioned time between sample collection and endpoint
- Storage = stored at -80 degrees of Celsius
o Not mentioned timing etc.
- Extraction = at 40 degrees of Celsius
The sample collection, handling, storage and analysis seems suitable for measuring omics data,
however limited information is mentioned.
Issues to take into account:
- Transport
- Laboratory errors
- Storage effects
- Freeze-thaw cycles
- Batch effects
You need a quality control sample
5. Did the authors use univariate and/or multivariate statistical methods for analyzing their data?
What are the differences and/or advantages of both approaches?
Therefore the author use univariate statistical methods, because there is only one outcome MAO
= yes/no logistic regression
However, it is a multivariate methods, because they are looking at all the metabolites together
- PLSDA
- Principal component analysis
- Repeated measures ANOVA
Complex data driven explorative
QUESTIONS
1. What is the research question of the study by Chen et al, 2015? Does this study focus on
prediction or on understanding? Explain.
The aim of the study is to identify important metabolites to distinguish a normal metabolic state
from an abnormal obesity metabolome.
Explore the differences between the MHO and MAO groups
The focus of the study is understanding, because they want to find metabolomes that identify
potential metabolic pathways that may regulate the different metabolic characteristics of obesity.
Etiologic research
o Understand mechanisms
o Aim to establish causality
But also prediction
2. Does this study examine a biomarker of exposure, susceptibility, or (intermediate) endpoint?
Explain.
The biomarker is an endpoint, because the biomarker reflect the type of obesity, namely
metabolically healthy obesity (MHO) or metabolic abnormal obesity (MAO) group. So with the use of
metabolites one distinguish MHO and MAO.
Also an exposure and susceptibility marker
3. Are the metabolomics measurements performed targeted or untargeted? Explain the difference.
What are the advantages and disadvantages of both approaches?
Both a targeted and untargeted analysis was employed in this study.
Target measurement = used to estimate the importance of each metabolite within a specific
pathway.
- Hypothesis driven
- The metabolites you want to investigate are already identified (specific)
- Aim = absolute quantification of specific metabolites in biological samples
- Limited number of metabolites
Advantages:
- Used when you want to investigate a specific metabolite
Disadvantages:
- Require a priori knowledge of metabolites of interest
- Do not achieve global coverage
Untargeted measurement = quantify the global metabolic profile of a sample
- Hypothesis-generation
, Practical I – Biomarkers: from single markers to omics
- Collect and analyse all detectable metabolites in a sample
- You do not know the identification of the metabolites
- Compare differences (peaks) in metabolic profiles across different (sub)populations
- As many metabolites as possible
- Relatively easy automated high-throughput
Advantages:
- Reveal novel compound and chemical unknowns
- Leading to new insights in disease mechanisms or treatment
Disadvantages:
- You do not know the identity of the metabolites
Library = database that contains information to identify metabolites
4. Was the sample collection, handling, storage and analysis suitable for measuring
(metabol)omics data? Why (not)? Which issues should be taken into account?
What should be mentioned in the method section:
- Sample collection = blood samples after 8 hours of fasting
o Not mentioned time between sample collection and endpoint
- Storage = stored at -80 degrees of Celsius
o Not mentioned timing etc.
- Extraction = at 40 degrees of Celsius
The sample collection, handling, storage and analysis seems suitable for measuring omics data,
however limited information is mentioned.
Issues to take into account:
- Transport
- Laboratory errors
- Storage effects
- Freeze-thaw cycles
- Batch effects
You need a quality control sample
5. Did the authors use univariate and/or multivariate statistical methods for analyzing their data?
What are the differences and/or advantages of both approaches?
Therefore the author use univariate statistical methods, because there is only one outcome MAO
= yes/no logistic regression
However, it is a multivariate methods, because they are looking at all the metabolites together
- PLSDA
- Principal component analysis
- Repeated measures ANOVA
Complex data driven explorative
