A microbiome is a collection of microbes, such as bacteria, fungi, protozoans and viruses, that inhabit a given environment. One such environment is the human body; the human microbiome is important for maintaining health, and when things go wrong it can contribute to disease.
The amount of data in a microbiome dataset depends on the location of the sample taken from, for example a human subject or environmental location under research. The graph shows differences in the number of species within a certain site of the human body. Gut and Skin microbiomes contain a higher number of species compared to those within the Respiratory and Urogenital as they are more exposed to many external environmental microbiomes from food, drugs, clothing and skincare product application.
When analysing large datasets, such as those from microbiome studies; there are some important questions to consider:
It is also important to keep a detailed analysis report of what tools, analysis methods and versioning were used.
Analysing microbiomes will yield a large amount of data, one thing to consider when designing an experiment
See below an example of how to set up and run a microbiome workflow:
Pipelines are often complex scripts, which are developed by experienced bioinformaticians. The key is to encapsulate and deploy the pipelines to the scientists/users, such that they can click & run pipelines with a simple interface. This workflow needs to also include easy steps for setting pipeline parameters.
Microbiome data analysis is still advancing, therefore it is difficult to select an exact technique to use. For the best results, it is important to use a diversity of techniques.
If using many different analysis techniques (starting with different assumptions) produces results that agree with each other, this will increase confidence in the results.
The answer to this question is dependent on the original purpose of the experiment or study.
The microbiome community has not decided on what type of techniques are better and there are many different techniques
Good advice is to go to PubMed and look for reviews of recent microbiome studies to keep informed about the techniques being used by other professionals.
Data needs to be organised and analysed, alongside the experimental techniques and surrounding metadata to give the datasets context.
Microbiome datasets bring together a lot of information such as experimental data, results and metadata; all this information needs to be managed effectively, to integrate all internal and external data.
The large size of microbiome datasets and their metadata makes it difficult to:
The FAIR data principles can help with this.
Example from https://www.eaglegenomics.com/mind-your-metadata/
Metadata serves a variety of purposes, with resource discovery one of the most common helping to manage and navigate these huge microbiome datasets.
Microbiome research is still a very developing field and there is so much explore, for example:
There is research being conducted to try and industrialise microbiome testing techniques, leading to products based on the microbiome for use in everyday personal and health care.
One resource that may help with understanding microbiomes: https://www.eaglegenomics.com/microbiome-knowledge-base/
Microbiome innovation will influence trilion-dollar markets across multiple industries. Organisations are recognising the potential of this new frontier for innovation.
The scope of the microbiome opportunity is huge, from skincare and drugs to nutrition and animal health - for companies with products that have an impact on health & wellbeing, it's very likely innovation from microbiome research can positively impact customers.
We're helping organisations gain insights from microbiome data to enable product innovation in consumer goods and healthcare.
To find out how our platform can support your microbiome research, please complete the enquiry form.