Genomics, Proteomics, and Metabolomics Will Unite To Complete Systems Biology
Schematic diagram depicting the relationship between genomics, proteomics, interactomics, and metablomics. Source: US Department of Energy.
According to a new review published in Nature Rheumatology,1 the wide and deep oceans of genomic, proteomic, interactomic, and metabolomic data that researchers have spent the last few years gathering and sorting are now ready to be used in a multidisciplinary and cross-pollinating fashion. Most scientists will balk at the prospect of taking vertically integrated systems biology into account during experiment planning due to how much it complicates investigations, but powerful new software and data analysis tools have paved the way for multi-omic research to be accessible to the average researcher. And this new study could prove extremely beneficial to human disease research if pursued.
The Nature Rheumatology review makes the case for combining metabolomics with genomics specifically to rheumatic diseases, which are typically linked to inflammation and by extension, metabolism.2 3 It’s especially tempting to combine genomic data and metabolomic data because genetic factors unambiguously underlie quantitative measures of the human metabolism, which could be used to predict disease progression or medication response—if you can crunch the data, that is. Since 2012, the number of characterized metabolites have exploded, leading researchers to the inevitable conclusion that each individual’s metabolome could be a more complex system than even their genome and that their graduate students will be cut off from pursuing projects of that scope.4 It’s easy to see why: Metabolomes, like genomes and proteomes, are self-altering dynamic equilibria that rely on both the genome and the proteome for their initial seed and subsequent reactions. Every metabolomic system impacts countless others, and is in turn altered as well, often in a self-regulating fashion.
Linking Experimental Data to Symptomatology
If metabolomes are more complex than genomes, what information will researchers need to keep track of in order to utilize metabolomic data in the context of researching disease? In a typical non-omic immunology study of a common disease like arthritis, the information required is relatively “simple”:
- Profile of disease symptomatology
- Profile of disease biomarkers
- Patient and control genomes
- Patient and control symptomatologies
- Patient and control sample metadata
- Patient and control samples’ measured disease biomarkers
- Characterized experiment-specific cellular or physiological systems
- Experiment-specific apparatus components
However, a typical study can only deliver typical quantities of data.5 In the above example, astute observers will note that unless the study is specifically a genetic study, the inclusion of genomic data is largely a formality ubiquitously performed for the sake of setting up further research in the event of an interesting result. In other words, the genetic data isn’t gathered with a concrete purpose in mind, so much as an insurance policy that interesting results can be replicated in a subsequent more-focused study cohort of a particular genetic profile that had the interesting result. This kind of experiment planning behavior will definitely continue, but it’ll be massively increased in scope, and have immediately relevant data to offer during each study’s first pass.
Linking Symptomatology to Genomics, Proteomics, and Metabolomics
Reflexively gathering metabolomic and proteomic data will become standard in the way that gathering genomic data is now, but that’s quite an undertaking even by an optimistic perspective. Each experiment will produce gargantuan amounts of data, with much of it left unused—until it’s time to start a new experiment based off of the harvested data.
To make the most out of matured systems biology models in the new research environment, even a formerly simple immunological study of a common disease needs to become an order of magnitude more complicated, requiring:
- All former experimental data
- Metabolomic disease biomarker profiles
- Genomic disease regulatory profiles
- Proteomic disease transcriptomic profiles
- Metabolomic data of samples
- Genomic data of samples
- Proteomic data of samples
- Proteomic profiles arising from genomic profiles
- Metabolomic profiles arising from genomic profiles
- Unique disease profiles of sample metabolomic profiles
- Unique disease profiles of sample proteomic profiles
- Unique disease profiles of sample genomic profiles
- Unique disease profiles of sample physiological profiles
- Unique disease profile resulting from summation of all aforementioned factors extrapolated forward and backward in time
It’s clear that this mountain of data will be extremely rich in clinical and experimental insights, provided that researchers can interrogate it systematically. Indeed, researchers will have to form workgroups which focus on specific aspects of each experiment’s massive amount of data, as it’s unlikely that any given researcher will be knowledgeable in disease pathology, genomics, proteomics, and metabolomics, not to mention yet-unexploited fields like interactomics. Instead of individual analysis, research groups who study diseases in a vertically-integrated ‘omic fashion will be nexuses of data sharing and intra-group as well as inter-group collaboration.
Systems biologists have known for a long time that collaboration is necessary to get the most data out of their field, and many software solutions exist for sharing various aspects of researcher data, yet no single all-encompassing platform exists.6 Data quantity is a major stumbling block for extant systems, as is seamless collaboration. Because researchers don’t use a comprehensive information technology solution to process their multi-omic data and share their assets with each other, collaboration efforts are bound to be plagued by the difficulty of transferring data between systems and struggling to get multiple research teams on the same page. For small data or single-omic studies of the past, the current informatics systems work fine—but their time is at an end with the rise of multi-omic research. Thankfully, a new and powerful software suite exists which can be used to manage and share the galaxy of multi-omic data.
Science Cloud is the collaboration, asset tracking, group analysis and experimental planning software that’s built to be a hub for all omics data and subsequent research. Using Science Cloud, you’ll be able to link genomic data sets to proteomic and metabolomic data, track sample metadata and share it all with collaborators. Working together to improve human health outcomes by performing vertically integrated ‘omic research is now possible. Contact us today to find out how you can use Science Cloud to jump into the future of mixed ‘omics.
- “Mixing omics: combining genetics and metablomics to study rheumatic diseases.” February 2017, http://www.nature.com/nrrheum/journal/v13/n3/full/nrrheum.2017.5.html ↩
- “What is a Rheumatologist?” April 2015, http://www.rheumatology.org/I-Am-A/Patient-Caregiver/Health-Care-Team/What-is-a-Rheumatologist ↩
- “The interplay between inflammation and metabolism in rheumatoid arthritis.” September 2015, http://www.nature.com/cddis/journal/v6/n9/full/cddis2015246a.html ↩
- “Innovation: Metablomics: the apogee of the omics trilogy.” March 2012, https://www.ncbi.nlm.nih.gov/pubmed/22436749?dopt=Abstract&holding=npg ↩
- “What difference does quantity make? On the epistemology of Big Data in biology.” April 2014, http://journals.sagepub.com/doi/abs/10.1177/2053951714534395 ↩
- “FAIRDOMHub: a repository and collaboration environment for sharing systems biology research.” November 2016, https://academic.oup.com/nar/article/45/D1/D404/2572060/FAIRDOMHub-a-repository-and-collaboration ↩