Mini-Review
How to Use Proteomics and Bioinformatics to Design Clinical Treatment
Dale Feldman*
Corresponding Author: Dale Feldman, Department of Biomedical Engineering, University of Alabama, Birmingham, Birmingham, AL 35294 USA
Received: May 11, 2020; Revised: June 29, 2020; Accepted: June 27, 2020
Citation: Feldman D. (2021) How to Use Proteomics and Bioinformatics to Design Clinical Treatment. Proteomics Bioinformatics, 3(1): 154-156.
Copyrights: ©2021 Feldman D This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Share :
  • 282

    Views & Citations
To design a clinical treatment, there needs to be clinical performance design constraints of what the solution has to be able to do (that current treatments cannot already do). Proving one treatment is statistically significantly better than another (particularly current treatment) does not mean it makes a clinical difference; in essence is the proposed treatment worth using instead of the current standard of care and therefore warrants further investigation. Part of the design process is determining the bioprocesses that need to happen in order to achieve the clinical performance design constraints. These bioprocesses are mostly controlled or can be measured by production and delivery rates of proteins (whose upregulation can be determined by proteomics and bioinformatics). Therefore, proteomics and bioinformatics provide key information in the design process, but need to be coupled with other information to design a viable clinical treatment. They can provide information on upregulation of proteins and pathways when a treatment meets the clinical performance design constraints or a specific bioprocess rate is achieved as well as when the treatment does not meet these design constraints. It can therefore help determine how to optimize the treatment to meet the design constraints (either clinical performance design constraints or bioprocess design constraints) as well as why a treatment works.

Keywords: Proteomics, Bioinformatics, Treatment design, Performance design constraints
REVIEW

Research studies, including proteomics and bioinformatics, use the scientific method to test a hypothesis (including showing increases in gene expression) related to a specific question. The results of the experiment are discussed in the context of the original question and whether the hypothesis was proven or not. The results are also compared to other studies looking at the same or similar questions. Therefore, for interpretation of the study, it will look at how it compares to previous studies, what are the ramifications of the results related to the original question, and what needs to be examined next; further test the hypothesis or test a different hypothesis. In the case of proteomics and bioinformatics, the results are typically what genes and pathways are upregulated; with interpretation on what this means. Hypotheses are typically provenbased on whether something leads to a statistically significant difference (or level of up-regulation) [1]. However, a statistical difference does not mean the difference is significant enough to matter for treatment [1]. Also, not being able to prove the hypothesis does not mean it is not true. A typical t-test that does not prove the treatments are statistically different at the 95% confidence level only has about a 50% confidence level that they are in fact the same. The actual difference is what matters, with statistics just showing how likely that difference is real.

For designing of treatments, a different approach is needed. Instead of starting with a question there is a problem and need statement. Instead of a hypothesis, there are performance design constraints of what the solution has to be able to do (that current treatments cannot already do). Experiments are done to determine if the solution meets the design constraint(s), which is similar to testing a hypothesis; in fact, a design constraint can be written as a hypothesis. The big difference is that design constraints are quantitative and hypotheses are typically not; with hypotheses not normally quantifying how big a difference is required [2]. This approach is also necessary to justify the need for the 
study. If the proposed treatment does not try to do something significantly clinically better (e.g. Meeting the clinical performance requirement), it is hard to justify the need for the study.

Implicit in the need statement is that it is a significant problem that is currently not being handled with current clinical treatment. Clinical performance design constraints are what a successful treatment needs to do. A successful treatment must be able to meet these clinical performance design constraints. Without a problem or performance design constraints, it is impossible to decide if the statistical difference shown actually makes a clinical difference; in essence is the proposed treatment worth using instead of the current standard of care and therefore warrants further investigation [2].

What this means for proteomics and bioinformatics is that information on upregulation of genes and pathways has to be put in a design context. Part of the design process is not only determining the clinical performance design constraints, but also the bioprocesses that need to happen in order to achieve the clinical performance design constraints [2]. Most of the time both of these are rates. For example, to achieve the desired clinical healing rate, there are needed bioprocesses rates as well (proliferation rate, migration rate, ECM production rate, etc.) [2]. These bioprocesses are mostly controlled or can be measured by production and delivery rates of proteins (whose upregulation can be determined by proteomics and bioinformatics) [2].

Therefore, proteomics and bioinformatics provide key information in the design process, but need to be coupled with other information to design a viable clinical treatment. They can provide information on upregulation of proteins and pathways when a treatment meets the clinical performance design constraints or a specific bioprocess rate is achieved as well as when the treatment does not meet these design constraints [3,4]. It can therefore help determine how to optimize the treatment to meet the design constraints (either clinical performance design constraints or bioprocess design constraints) as well as why a treatment works [5,6].

In order to explain this better, a specific example will be used: Fracture healing of long bones in professional athletes, which require a bone plate for repair [7]. The problem is (1) The high complication rates and (2) That the designs interfere with healing; lengthening the rehabilitation time [7,8]. Many of the complications (e.g., refracture of the bone) can be reduced by speeding healing. In clinical practice, implants are removed (80% of the time in many cases) to speed healing and reduce long-term complications [7,8]. This typically requires a second rehabilitation cycle and, in many cases, leaves holes in the bone, which increase the susceptibility to refracture [7,9]. The clinical performance goal is to return the athlete to the activity as soon as possible. Clinical performance design constraints (in comparison to the current standard of care) could be to reduce the total healing and recovery time by 50% and the overall cost of treatment by 25% [2,7,8].

Proteomic and bioinformatics studies can determine the proteins and pathways upregulated throughout the healing process that lead to successful (or unsuccessful) treatments (i. e. meeting the clinical performance design constraints) [3-5]. This helps determine the desired bioprocess rates and potential strategies for meeting these bioprocess rates as well as screening potential treatment options [1,3-6]. These studies help show that meeting the clinical performance design constraints requires an increase in healing rate by at least 20%; eliminating the need for a removal surgery (with a second rehabilitation cycle) [2,6-7]. They also show which genes and pathways are upregulated when the healing rate is increased [3-6]. These studies can also look at different strategies that can affect healing and why some are more effective than others. In addition, they show why different implementations of each strategy can lead to different outcomes in an effort to help optimize each potential strategy [3-5]. They can also help show why an increase in loading on the fracture site is an effective way to meet the 20% increase in healing as well as why certain loading strategies work better than others [2,7]. Further, they can help show how a reduction in stiffness of the bone plate leads to upregulation of the key genes and pathways necessary to speed the healing, predominantly by increasing the loading on the fracture site [2,7]. In addition, they can help explain why the stiffness of the bone plate needs to decrease over time to meet the 20% increase in healing rate [2,7]. They also can help show why there is a maximum for the amount of loading, on the fracture site, at any given time and occurs below the level of refracture [2,7,10].

CONCLUSION
Therefore, these studies can help show that a degradable metal bone plate would meet all the design constraints as well as help determine the optimal range of degradation rates for the bone plate, plus what initial properties are needed [2,7]. They also can help show that a magnesium metal bone plate degrades too quickly and needs to be modified to slow the rate [2,7,11].

So, proteomics and bioinformatics provide key information on why one strategy is better or worse than another plus help show how to tweak these strategies to meet the performance design constraints. Although it is straightforward to determine if a treatment meets the performance design constraints, to optimize a design the bioprocess rates necessary to meet these performance design constraints need to be determined. Proteomics and bioinformatics can help determine why treatments have different bioprocess rates as well as why having certain combinations of bioprocess rates move closer or further away from meeting the clinical performance requirements. Understanding these relationships is important in tweaking a treatment protocol to meet the performance design constraints. Otherwise, it is difficult to predict how a change in the treatment would lead to a significant change in clinical performance.
  1. Feldman D (2017) An engineering approach to the scientific method. Sign Bio Eng Bio Sci 1(1).
  2. Feldman D (2019) Quantification and modeling of biological processes for tissue engineering and regenerative medicine. Biomed J Sci Tech Res 12(5): 1.
  3. Jennings A, Chen D, Feldman D (2010) Upregulation of chemokine (C-C motif) ligand 20 in adult epidermal keratinocytes in direct current electric fields. Ach Dermatol Res 302(3): 211-220.
  4. Jennings A, Chen D, Feldman D (2008) Transcriptional response of dermal fibroblasts in direct current electric fields. Bio Electromag 29(5): 394-405.
  5. Stoff A, Rivera AA, Mathis JM, Moore ST, Banerjee NS, et al. (2007) Effect of adenoviral mediated over expression of fibromodulin on human dermal fibroblasts and scar formation in full-thickness incisional wounds. J Mol Med 85(5): 481-496.
  6. Feldman DS, McCauley JF (2018) Mesenchymal stem cells and transforming growth factor-β3 (TGF-β3) to enhance the regenerative ability of an albumin scaffold in full thickness wound healing. J Funct Biomater 9(4): 65.
  7. Sealy M, Ziye L, Chao L, Guo Y, White B, et al. (2017) A strategy to optimize recovery in orthopedic sports injuries, J Bioanal Biomed 9: 144-151.
  8. Parker PM (2008) The 2009-2014 world outlook for medical and surgical bone nails, plates, and screws and other internal fixation devices, pp: 16-17.
  9. Grimm MJ (2009) Biomedical Engineering and Design Handbook. McGraw-Hill, New York, pp: 421-444.
  10. Sealy M, Guo Y, Caslaru R, Sharkins J Feldman D (2016) Fatigue performance of biodegradable magnesium–calciumalloy processed by laser shock peening for orthopedic implants. Int J Fatigue 82: 428-436.
  11. Kirkland NT, Birbilis N (2014) Magnesium biomaterials: Design, testing and best practice. Springer International, Cham, Switzerland.