The number of published scientific papers on the microbiome has increased year after year for the past decade, and is now at an all-time high. Research is moving beyond preliminary association studies, toward more sophisticated next-generation sequencing analyses that will underpin novel diagnostics and therapeutics.
Yet when you take a close look at the data analysis methods from published microbiome papers, you find huge variation in all parts of the workflow: sample collection, DNA extraction, sequencing, and bioinformatics. The workflow in every lab may be slightly different depending on the physical setup as well as the available personnel, budget and other resources. This means when you find something novel and exciting in the preliminary data analysis for your microbiome study, you may wonder whether you’re seeing a truly interesting finding that adds to what previous studies have demonstrated, or whether the result is an artefact of some aspect of your workflow. For example, a higher or lower concentration of bacteria may have been observed not because it’s actually in the sample, but because of how the sample was treated. In two large-scale microbiome projects, the MicroBiome Quality Control project (MBQC) and the International Human Microbiome Standards (IHMS), DNA extraction protocols were observed to have the largest impact on the variability of results from one experiment to another. Other parts of the workflow, however, must not be overlooked.
The problem, in theory, could be fixed by standardizing microbiome analysis methods and equipment across the field.
A multitude of [microbiome research] consortia on a regional and global scale have been formed,
said life sciences executive Christiane Honisch, in a webinar called Toward Standardized Workflows.
The standardization of workflows becomes an integral part of data comparability across the research community.
Standardization would mean more accurate, high-quality lab results that are easily comparable to results from other labs globally; it would also mean less repetition of analyses due to issues in the workflow, and more efficient training of lab members.
But widespread standardization is easier said than done while technologies are still improving and protocols are still being optimized for different circumstances. Many groups (for example, here) are aware of—and making progress toward—the goal of standardizing microbiome analysis. However, those in the field are still a long way from agreeing on how to achieve the highest-quality results for these complex analyses.
Without globally-accepted standards to lean on, what can a researcher do in order to maximize the chances that a microbiome study will yield high-quality results that are reproducible and comparable to related studies?
“Best practices are critical,”
noted Rita Colwell, University of Maryland Professor and life sciences executive, in the Toward Standardized Workflows webinar.
The following tips can help you move toward best practices in order to maximize the reproducibility and comparability of your next microbiome study:
1. Learn about the protocols and supplies used in larger research projects in your area of interest.
Familiarize yourself with the methods used in projects carried out by large studies or research consortia that have samples being analysed in multiple labs. Better yet, talk with those who were involved in the front-line lab work and ask them about any particular challenges they encountered. You may decide to use the information to shape your own analysis—which could give you more confidence that your results will be comparable to large, landmark studies.
2. Consider automating part of your workflow to reduce human error.
Is there a part of your workflow that’s especially difficult to keep consistent based on your particular lab setup? Perhaps your space lacks climatization or has people cycling in and out who might need to move equipment or change settings. Consider investing in automation equipment for a specific part of the workflow in order to reduce or eliminate that source of error. If your budget doesn’t allow for your own automation equipment, you might team up with a nearby group to purchase and share the equipment. Strategically automating even one part of your workflow could go a long way toward increasing your confidence that your results are reproducible—and it could even lower stress and communication / training difficulties for those working in your lab.
3. Select equipment and supplies typically used by other labs, and do your research on the quality control measures of the company you are ordering from.
For reproducible and comparable results, choose high-quality equipment and consumables from trusted companies that ship globally. This is an easy way to eliminate one area of potential difference between your project and other projects. For budgetary reasons it may be tempting to order lower-quality equipment, especially if it ships quickly, but quality equipment lowers the chances of malfunction and might just save the expense of having to re-do the study entirely.
Before ordering lab supplies and equipment from a vendor, take a few moments to search the company website and learn about the quality control measures. Find out what international standards the equipment adheres to, and what you can expect when you use it.
Also, be sure to keep your equipment serviced and maintained so it continuously operates within the parameters needed to produce high-quality results. Being proactive is far better than having your equipment malfunction at a crucial moment for your experiment. Check the manual or email the sales representative if you have questions about proper maintenance of lab equipment.
4.Stay organized and efficient by streamlining how your lab keeps track of samples and organizes / accesses data.
Lab communication and organization can pose challenges as you gain new lab members or if you must implement restrictions on the number of individuals working together in the space. An electronic lab journal can greatly improve efficiency and reduce misunderstandings by keeping samples, protocols, and data in order. Another major benefit of digitising your data is the ease with which you can share information and results with your collaborators. Using these tools, you can assign clear roles and permissions, while features such as Sign & Witness help you comply with GLP guidelines. Overall this ensures that nothing is overlooked and that your data quality is as high as possible.
5. Use appropriate controls to screen for contamination and allow for data comparisons on a global scale.
Controls provide reassurance that your results are of a high quality and that you are picking up a real signal rather than contaminants. Controls to consider for your analysis include:
- Negative template control for reagent and laboratory contaminants
- Positive template control for assay validity and laboratory contaminants
- Internal spike-in control for assay validity, quality of sequencing preparation and instrumentation, as well as normalization and stratification
Following these suggestions may increase the chances that you find the ‘ground truth’ about your samples and make a meaningful contribution in the microbiome field. Equipped with these and other best practices, you can use your available funding to increase the number of samples you analyze, rather than having to re-do your analyses to get more accurate results.
Laboratory equipment, supplies & services
Visit us: www.eppendorf.com