Finding the solution to a problem using DNA isn't as easy as using a standard computer by a long shot. While DNA computing may one day be automized and simple as inputting a problem and pushing a button, like a regular computer, today it requires a lot of patience, lab work, and time. This section will outline the main steps used to eventually solve our 5-city problem. The sections include: Finding Possibilities, Correct Starting and Ending Destinations, Correct Number of Cities, and Cities Appear Only Once
For our DNA computing problem, our 5-city, problem, we should first generate all possible solutions to the problem, so that we can continually narrow our possibilities with special techniques to find the answer later on.
To find all possible paths between cities using our connection rules, we use a machine called a DNA synthesizer to make our own partial DNA strands with predefined letters. In our case, we would create codes for each city first. Let's just say Cincinnati is ATCGGC, Columbus is GTACGA, Dayton is GCCGTA, Cleveland is GCATTA, and Toledo is CGTAGC. There, now we have coded all the cities in our problem. Now, we code in the connections between our cities. We can do this by taking advantage of the fact that DNA bases bond with other bases. For example, lets code in the connection from Dayton to Columbus. To do this, put the two codes for these cities (Dayton first) side by side. You would get GCCGTAGTACGA. Now, take the middle six bases(or the last three in Dayton followed by the first three in Columbus) and find their complement (the DNA strand that it would pair with, remember C and G go together and A and T go together). This would be CATCAT. Now, if these three strands of DNA were mixed together, the last three letters in Dayton would connect to the first CAT and the first three letters in Columbus would connect to the second CAT. See the chart below for all the connections. Now, all we have to do to find the connections is mix a large number of these partial strands (like 10 raised to the fifteenth power, to make sure we find all possible solutions, remember, making this amount of DNA is not hard because of its replication capabilities) together until they all connect using a substance called ligase, forming all possible paths using our cities and our connection rules.
So now we have all the possible city routes. The first step in the road to narrowing our possibilities down to one is to select DNA with the correct starting and ending cities, Cincinnati and Toledo.
To do this we take advantage of a special enzyme known as polymerase. This enzyme will selectively copy part of a single strand of DNA over and over again, using a primer (a primer is the compliment of a short DNA strand) as a starting point. By using Polymerase Chain Reaction, as the process is called, we can place two primers, one for the starting base of Cincinnati and one for the ending base of Toledo, and therefore copy strands that start with Cincinnati and end with Toledo over and over again. After the Polymerase Chain Reaction is finished, we end up with routes that begin and end with the correct cities.
The next step in the elimination process is to narrow our search down to DNA strands with the correct number of cities, which is five, since each city must be passed through exactly once.
To do this, we run the DNA through a gel matrix, full of interlocking polymer that forms loops that the DNA strands that we have must weave their way through. The DNA run through this matrix because they are usually negatively charged, so will be attracted to a postivie source that they gel matrix provides on one end. Because the DNA must weave its way through, shorter strand move faster. After the matrix has been run, there is typically a number of DNA bands, each representing a length. Since each city is six bases long, and there are five cities, we simply take out the bands that are made up of 30 bases. We now have all routes that have the correct number of cities.
The last step in this process of elimination is to find solutions that have each city only once in them.
This is actually fairly easy. Using a process called affinity purification, we can put a sort of ball in the strands of DNA with a tag on it that can hybridize with a certain city. For example, If you put a ball in with a tag that would hybridize with Cincinnati, a possibility with Cincinnati would in it would attach. By mixing these balls of a certain city in and then putting balls with different cities in each time, we can find the one route that has five cities, the correct starting and ending cities, and each city used only once. These fit all the requirements of our original problem (see DNA Computing Example), so now, all we have to do is read out the bases, convert them to cities/routes, and we have our answer, Cincinnati to Columbus to Dayton to Cleveland to Toledo! You have just solved a problem using DNA!
Introduction
Are You Really A Computer?
DNA Computing Example
Finding Solutions
The Future
Games for DNA Computing
Genetics - DNA Computing - Are You A Computer?