New research from scientists at Lenovo, Yale, and Johns Hopkins University reveals the recipe for a revolutionary treatment in the form of a cardioprotective biochemical cocktail. The discovery of using factors secreted by stem and stromal cells in normal intercellular communications to harness the protective and restorative effects on human tissue is an important breakthrough. Until now, capturing and mapping secretory dynamics data was a major obstacle to mastering and managing this “key component of therapeutic recombinant cocktails for tissue repair,” say the researchers. A new method for reconstructing the data solves these issues and suggests more potential for computing in other use cases despite the absence of key data.
The discovery
Previous researchers established that intercellular communications, specifically paracrine signaling, is the likely keystone in the therapeutic potential of adult stem and stromal cells in disparate disease presentations. Stromal cells were known to change their secretory signatures to trigger a specific healing response in injured tissue.
In general, there are four main forms of cell signaling used in intercellular communications. The first is autocrine signaling in which the cell releases factors that bind to receptors on the same cell. The second is direct contact, aka juxtacrine signaling, wherein the cells are literally in direct contact and factors are sent from one cell to another through gap junctions in the cell walls. The third is paracrine signaling which means a release of molecules (factors) by one or more cells into extracellular space where they cross a short distance before connecting with cells that have receptors that fit those specific molecules. The fourth is endocrine signaling which works the same way but over greater distances such as via the bloodstream.
This new research focused on mimicking the dynamics and dosage of stromal cell-derived secreted factors in paracrine signaling in a cocktail to elicit tissue repair after a myocardial infarction (MI) event, but similar use cases can repair other tissue injuries such as the brain after a stroke.
“If factors secreted by adult stem and stromal cells can indeed have tissue-protective and therapeutic effects, mimicking their function constitutes a new cell-independent treatment modality,” wrote the researchers in .
Reconstructing missing key data
While these findings constitute significant medical implications, they were not outputs of a standard bioinformatics project. Key data was missing and without those inputs, an algorithm could produce an inaccurate or even wrong conclusion. However, obtaining the missing data wasn’t just a matter of gathering it through typical technological means or buying it from another source. The data simply did not exist and consistently evaded capture.
Two major challenges made collecting precise measurements and mapping cellular changes at the point of instigation virtually insurmountable: technical limitations in analyzing complex solutions and capturing data from a dynamic and highly adaptive cell microenvironment.
The researchers solved these challenges by reconstructing secretion dynamics using spatial information. By using an array of signal detectors at various distances from the secreting stem cell source, they found that “the inverse problem of reconstruction of the signaling dynamics from the resulting spatial profile can still be solved even for such more complex cases.”
In short, an algorithm reconstructs the continuous dynamic profile from a few spatial profiles.
The wider implications
The success of this effort suggests that spatial and other data can be used to reconstruct missing data for other computations in medical treatments and additional use cases. In other words, it may not be necessary to capture data directly where obstacles seriously impede or prevent such measurements. Instead, other data can be used to reconstruct information accurately in place of direct data reads.
Bioinformatics and HPC have proven invaluable to computing some of the most complex problems in Life Sciences today. However, using huge data sets was once the biggest challenge to researchers, today filling the gaps in data is fast becoming a larger obstacle. Fortunately, today there are several options in finding the missing information and inferring more precise answers. Those options include simulated data, enhanced data, and reconstructed data.
In short, analysis is now possible using increasingly disparate data sources some of which are improved beyond their original versions with various enhancements, reconstructions, and distillations. This means that digitalization and complex computing can exceed the limits of existing data sets.
HPC potential expands
The possibilities in seemingly endless manifestations of data further HPC’s potential to calculate the previously incalculable and to compute the previously unknowable.
A Hyperion report finds that $160 in revenue is generated for every $1 invested in HPC in life science. Such high rates of return on investment (ROI) are attributable to several factors, most notably through greater innovation, process optimization, and new R&D opportunities.
Lenovo has a strong record in helping many of the world’s best research teams leverage HPC and bioinformatics to conquer some of the most complex challenges mankind has ever faced and improve the human experience.
Calculating these massive data sets takes a supercomputer, and Lenovo is the industry leader in High Performance Computing (HPC), with over one third of the TOP500.org supercomputer list being Lenovo branded systems. Additionally, Lenovo has supercomputers in more countries than any other supplier does. Our customers are on the forefront of research in genomics, precision medicine, diagnostics and drug development. They trust Lenovo to deliver the performance needed to drive these and other research projects that help solve humanity’s greatest challenges.