* -35 box of the promoter <pre>... <minus35box>...</minus35box> ...</pre>

* -35 box of the promoter <pre>... <minus35box>...</minus35box> ...</pre>

** First, I knew I needed to identify the sequence that I'd be looking for within the file. The week 4 assignment indicated that the consensus sequence for the -35 promoter sequence is <code>tt[gt]ac[at]</code>. In thus, I knew I needed to plug this sequence into <code>sed</code> in order to filter for this sequence. Because I wanted a single replacement of one sequence, I knew that <code>sed s//g</code> would be the best option. My first theory was to try for <code>sed s/tt[gt]ac[at]/ & /g</code>, to put a space on either side of the sequence. This would test whether or not it was finding the sequence correctly, before I put in the tag.  

** First, I knew I needed to identify the sequence that I'd be looking for within the file. The week 4 assignment indicated that the consensus sequence for the -35 promoter sequence is <code>tt[gt]ac[at]</code>. In thus, I knew I needed to plug this sequence into <code>sed</code> in order to filter for this sequence. Because I wanted a single replacement of one sequence, I knew that <code>sed s//g</code> would be the best option. My first theory was to try for <code>sed s/tt[gt]ac[at]/ & /g</code>, to put a space on either side of the sequence. This would test whether or not it was finding the sequence correctly, before I put in the tag.  

** I tested using the command <code> cat infA-E.coli-K12.txt | sed "s/tt[gt]ac[at]/ & /g" </code> However, this command did not work, since it changed every single one that appeared, not just the first! Since I only wanted the first one to be changed, I did some research to find out how to change the first iteration using sed. Using [http://unix.stackexchange.com/questions/155805/sed-replace-first-k-instances-of-a-word-in-the-file this link from Stack Overflow], I learned that the /g in the command was indicating to change every single iteration. Changing it to /1 would cause it to change only the first iteration! Running <code> cat infA-E.coli-K12.txt | sed "s/tt[gt]ac[at]/ & /1" </code> resulted in the output I expected. As a result, all that was left was to find the first and last space and replace then with the starting and ending tags.  

** I tested using the command <code>cat infA-E.coli-K12.txt | sed "s/tt[gt]ac[at]/ & /g"</code>. However, this command did not work, since it changed every single one that appeared, not just the first! Since I only wanted the first one to be changed, I did some research to find out how to change the first iteration using sed. Using [http://unix.stackexchange.com/questions/155805/sed-replace-first-k-instances-of-a-word-in-the-file this link from Stack Overflow], I learned that the /g in the command was indicating to change every single iteration. Changing it to /1 would cause it to change only the first iteration! Running <code> cat infA-E.coli-K12.txt | sed "s/tt[gt]ac[at]/ & /1" </code> resulted in the output I expected. As a result, all that was left was to find the first and last space and replace then with the starting and ending tags.  

** However, this ended up being harder than expected. Because </minus35box> had a / key, sed interpreted that as the end of the input. The forward slash would need to be escaped in order for sed to treat it not as a part of the command, but rather as a string. I knew that in other command line arguments, a backslash placed before the offending character would escape the character, allowing it to be read as a character. This held true for the sed command as well. The final command and output was as follows:  

** However, this ended up being harder than expected. Because </minus35box> had a / key, sed interpreted that as the end of the input. The forward slash would need to be escaped in order for sed to treat it not as a part of the command, but rather as a string. I knew that in other command line arguments, a backslash placed before the offending character would escape the character, allowing it to be read as a character. This held true for the sed command as well. The final command and output was as follows: