Team:Jefferson VA SciCOS/Results

From 2014hs.igem.org

(Difference between revisions)
 
(6 intermediate revisions not shown)
Line 1: Line 1:
<div id="header">{{Team:Jefferson_VA_SciCOS/Header}}</div>
<div id="header">{{Team:Jefferson_VA_SciCOS/Header}}</div>
 +
 +
{|align="justify"
 +
|
 +
'''''<h3>Results and Conclusions</h3>'''
 +
----
 +
 +
'''''Testing''' <br />
 +
We hoped to run two gels to see if our new node and composite pathway were made properly. We also planned to test our pathway and check the fluorescent properties of our node by transforming both in E.coli, but due to lack of time and resources, we were unable to determine definitively whether our efforts were successful. If we did have the time however, we would have followed this protocol to determine a false positive to true solution ratio in order to determine the efficiency of our bacterial computer in solving the Traveling Salesman Problem. <br />
 +
----
 +
'''''Screening for all correct phenotypes''''' <br />
 +
If some cells do not contain antibiotic resistance or do not fluoresce a certain color under UV light (bright white in this case), then we know that they do not represent a solution to the Traveling Salesman Problem. However, with this comes the possibility of a false positive solution.
 +
----
 +
'''''Using PCR to separate false positive solutions from true solutions''''' <br />
 +
In the case of false positives, all the gene halves are properly aligned to produce fluorescence – the first half is followed by the second. Yet the resulting path is not logical because it requires moving between nodes without following an edge. It has already been established that the length of a pathway that solves the Traveling Salesman Problem is the length of all split genes, and that the first and last node of a pathway is never flippable. Thus, PCR primers can be designed for the first and last node, and running the PCR products on a gel should give the lengths of the amplified DNA. If the path is false-positive solution, then the resulting bands should theoretically be too small to be the most optimal solution to the Traveling Salesman Problem.
 +
----
 +
'''''Detecting Fluorescence Using Spectroscopy''''' <br />
 +
To count the number of colonies that fluoresced a certain color on a plate, we would have used a fluorometer. In order to use the fluorometer for our construct, we would have first run scans to find the optimal excitation and emission wavelengths for the split BFP genes that we created. Once these wavelengths have been determined, the fluorometer could be used to determine the emission wavelengths of the fluorescent colonies and we could compare them to the previously determined optimal results. Depending on the number of colonies, they could be counted either by hand or by taking pictures and analyzing them through a program such as ImageJ. By compiling the results of all these different methods, it would be possible to determine the number of fluorescent colonies, how many of them were true solutions compared to false positives, and ultimately achieve our goal of evaluating the efficiency of our system.
 +
----
 +
'''''Conclusion''' <br />
 +
We successfully designed and created a 4-node pathway with varied ribosome binding ability in order to solve the Traveling Salesman Problem, but we were ultimately unsuccessful in implementing the pathway by testing its efficiency. However, this proof-of-concept experiment has the potential to demonstrate the applicability of synthetic biology to the solving of NP-complete problems, and could validate synthetic biology as a feasible approach to computing in the future. In order to demonstrate its feasibility to solve a wide range of problems in theoretical computer science however, various constructs will need to be developed that are designed to simulate other common problems, such as the clique problem and the rank coloring problem. Furthermore, it is recommended that in the future, more time be allotted to troubleshooting. Although we spent significantly more time for experimentation for this year’s experiment than we did for last year’s, we ran into several problems with the transformation and had to repeatedly fix certain variables of our procedure so that we ended up with a favorable result. To circumvent this issue, we recommend future researchers to be completely aware of the unique resources and environment within their lab so that they can adapt their protocols more easily. In addition, since we worked in a school laboratory, finding sufficient time to finish certain procedures was a bit of a hassle. For most of the year, we were only able to work during 45 - 1 hour 30 minute blocks, which is not enough time to complete a substantial portion of experimentation. To work around this issue, we recommend that future groups truly work as a team - map out everyone’s schedules and find out who is available when. Not everyone needs to be available for a procedure to get done; as long as one person can be in the lab, the job can be finished.
 +
|}
<html lang="en">  
<html lang="en">  

Latest revision as of 01:38, 21 June 2014

Results and Conclusions


Testing
We hoped to run two gels to see if our new node and composite pathway were made properly. We also planned to test our pathway and check the fluorescent properties of our node by transforming both in E.coli, but due to lack of time and resources, we were unable to determine definitively whether our efforts were successful. If we did have the time however, we would have followed this protocol to determine a false positive to true solution ratio in order to determine the efficiency of our bacterial computer in solving the Traveling Salesman Problem.


Screening for all correct phenotypes
If some cells do not contain antibiotic resistance or do not fluoresce a certain color under UV light (bright white in this case), then we know that they do not represent a solution to the Traveling Salesman Problem. However, with this comes the possibility of a false positive solution.


Using PCR to separate false positive solutions from true solutions
In the case of false positives, all the gene halves are properly aligned to produce fluorescence – the first half is followed by the second. Yet the resulting path is not logical because it requires moving between nodes without following an edge. It has already been established that the length of a pathway that solves the Traveling Salesman Problem is the length of all split genes, and that the first and last node of a pathway is never flippable. Thus, PCR primers can be designed for the first and last node, and running the PCR products on a gel should give the lengths of the amplified DNA. If the path is false-positive solution, then the resulting bands should theoretically be too small to be the most optimal solution to the Traveling Salesman Problem.


Detecting Fluorescence Using Spectroscopy
To count the number of colonies that fluoresced a certain color on a plate, we would have used a fluorometer. In order to use the fluorometer for our construct, we would have first run scans to find the optimal excitation and emission wavelengths for the split BFP genes that we created. Once these wavelengths have been determined, the fluorometer could be used to determine the emission wavelengths of the fluorescent colonies and we could compare them to the previously determined optimal results. Depending on the number of colonies, they could be counted either by hand or by taking pictures and analyzing them through a program such as ImageJ. By compiling the results of all these different methods, it would be possible to determine the number of fluorescent colonies, how many of them were true solutions compared to false positives, and ultimately achieve our goal of evaluating the efficiency of our system.


Conclusion
We successfully designed and created a 4-node pathway with varied ribosome binding ability in order to solve the Traveling Salesman Problem, but we were ultimately unsuccessful in implementing the pathway by testing its efficiency. However, this proof-of-concept experiment has the potential to demonstrate the applicability of synthetic biology to the solving of NP-complete problems, and could validate synthetic biology as a feasible approach to computing in the future. In order to demonstrate its feasibility to solve a wide range of problems in theoretical computer science however, various constructs will need to be developed that are designed to simulate other common problems, such as the clique problem and the rank coloring problem. Furthermore, it is recommended that in the future, more time be allotted to troubleshooting. Although we spent significantly more time for experimentation for this year’s experiment than we did for last year’s, we ran into several problems with the transformation and had to repeatedly fix certain variables of our procedure so that we ended up with a favorable result. To circumvent this issue, we recommend future researchers to be completely aware of the unique resources and environment within their lab so that they can adapt their protocols more easily. In addition, since we worked in a school laboratory, finding sufficient time to finish certain procedures was a bit of a hassle. For most of the year, we were only able to work during 45 - 1 hour 30 minute blocks, which is not enough time to complete a substantial portion of experimentation. To work around this issue, we recommend that future groups truly work as a team - map out everyone’s schedules and find out who is available when. Not everyone needs to be available for a procedure to get done; as long as one person can be in the lab, the job can be finished.