Assessment of study quality usually aims to develop a tool which is universal for all disciplines using a specific study methodology. However, the implementation of new methods may require further criteria to ascertain the validity of results - Which domains should be additionally addressed by observational studies investigating the associations between metabolomics and health outcomes?
Interesting question, Jakub Morze, do you intend to develop such a tool or improve an existing one? If so, could you reference an example or more detailed description of what is considered reasonable in other disciplines/which questions are usually adressed/how evaluation is done.
In my opinion, you have to consider 3 major sources of bias: experimental, technical and data analysis related points. Following a brief description of what I mean:
[experimental] everything up to sample storage in N2; including: appropriate selection of matched control subjects, balanced study design, sufficient size of the study population, consistent/reproducible sampling protocol, excellent documentation of meta data, clear communication of study limits (if only young, male white subjects are investigated you obviously can not make claims for the world population...), and many more
[technical] everything from sample preprocessing until MS measurement; smart sample randomization during extraction and measurement, reproducible protocol, additional QC samples; documentation and communication of meta data (extraction group, sample batch, runorder etc.) to investigate biases posteriori, documentation of machine quality (e.g. sample blanks, mass precision after tuning)
[data processing] everything following MS analysis; documentation and communication of workflow - ideally in reproducible way
All 3 groups are demanding in their own. To define formal criteria is hardest for the last step as one can set so many thresholds and parameters at multiple points in a pipeline that may change results drastically. You may want to ensure that reported results are rigorously tested for statistical significance and pipeline is documented and reproducible at least.
Let me know what you think. Bests, Jan
Hi,
Excellent question, and above all great in puts and excellent pointers from the #metabolomics side from Dr. Jan Lisec on an array of topics from limitations in global population-scale omics studies !
However, I would highly recommend you to consult and contact/ discuss with Dr. Karsten Suhre (http://qatar-weill.cornell.edu/research/research-faculty/suhre-lab) for insights on the interface of metabolomics and epidemiology/ observational/ nutritional towards MWAS-style studies! He is immensely helpful and interactive too! : )
Best regards,
Biswa
In addition, you should consider stability of targeted metabolites in your specific matrix and also potential matric effect are also essential and avoid numerous bias
Hi, This is a very interesting and relevant question for the entire research community working in metabolomics. Please feel free to register (for free to create a login with your Email ID!) and join the International Metabolomics Society’s Forum: http://www.metabolomics-forum.com/index.php?action=forum;?action=forum#c12 to ask more specific questions any software, database, approach and platform (mass spectrometry, NMR) issues where you can get very specific responses from expert users and researchers! Thanks again, Biswa
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science
- Reasoning