For pairwise alignment, you need two sequences which you expect to have a relationship. In the previous section on dotplots we used two sequences and we will continue with the same ones here. If you haven't already got the sequences open click here.
With the two sequences selected, click the Align/Assemble button on the tool bar. Select Pairwise Align. This will bring up a window that should look like this:
There are a variety of alignment methods available in Geneious Prime. This tutorial uses the Geneious Alignment function. You should reset any changed parameters to their defaults by choosing Reset to Defaults under the Settings cog at the bottom left, unless it is greyed out in which case you already have the defaults selected.
Click OK and the two sequences will be aligned producing a new document. This document will automatically be selected and should look similar to the screenshot below (you may have to turn off the annotations and choose to Wrap the sequences in the sequence viewer options to see the whole alignment more easily).
In addition to the alignment, you also have a dotplot for the two sequences. What you should notice is that the alignment doesn't look very good in places. There are few gaps and there are regions of high mismatch. This is happening because the default gap penalties are strict and the scoring table is too relaxed. In addition, a pairwise alignment does not deal with sequence inversions.
Task 1: In the Alignment View locate the first region of mismatches in your alignment view. Turn on the Consensus sequence and select Highlighting to make this task easier. These options can be found in the Display panel on the right hand side of the Alignment View. In the Highlighting options select Disagreements to Consensus from the dropdown boxes. With these options selected the bases in the alignment which match the consensus sequence will be grey but mismatches will be shaded (mismatches will either appear as black or colored depending on the zoom level on the alignment). Gaps will also be shaded unless you select Ignore gaps in the Consensus options. Mismatching regions will now have a striped effect making them easy to locate. When you have located the first region of mismatches, identify the start and stop coordinates for each sequence and locate that region on the Dotplot.
Note how this region of the alignment relates to what you see in that region of the dotplot. The first region of mismatches runs from 443-609 in the original sequence and 511-660 in the mutated sequence. The dotplot shows that this region has some mismatching and then some repeating sequence. Since there are multiple possible paths through this region, the aligner has used the mathematically least costly path but this does not reflect the true relationship since there is clearly a match to be made as indicated in the dotplot.
This task highlights one of the issues with sequence alignments. The alignment will follow the mathematically optimal path through the two sequences and that does not necessarily reflect what has really happened. It is cheaper for the alignment algorithm to force its way through with mismatches than to take the cost of a high gap penalty.
You should realign these sequences with modified settings. Rather than select the original two sequences, you can just select the previous alignment document and select Align/Assemble→Pairwise Align again as a shortcut.
The gap open and extension options are available under the More Options button. The gap extension penalty is set to 3 by default and this will limit the distance that a gap can be extended because every gap extension will subtract something from the overall score. Eventually, a gap will become too expensive compared with just accepting a large number of mismatches and this is exactly what happened with the previous alignment. Another problem is that the default scoring table favours mismatches.
To solve these two problems, change the default cost matrix to 93% similarity and change the gap extension penalty to 0, then click OK.
If you look at the alignment now with the same consensus and highlighting options set, you will see that there are almost no disagreement between the two sequences. If you uncheck Ignore gaps it will be clear that the aligner has now used gaps as a way to avoid mismatches.
In the Dotplot there is an inverted region. Locate this region on the new alignment. Remember, a reversed diagonal indicates a sequence inversion event. This means that one sequence has to match against the reverse complement of another sequence. To see the inverted region you will need to select the option Reverse Complement in the dotplot options. Note what the alignment algorithm has done here - in the inverted region, the alignment has simply misaligned the whole section. Pairwise alignments always run in a single direction so for inverted regions you have to be prepared to realign as you will see in the next task.
Task 2: In order to align the inverted region, it is necessary to select just those parts of the sequence. You can do this by clicking on the start of the region and dragging your mouse to the end. Make sure you select the two sequences otherwise you will not be able to extract the alignment region. You selection should look like this:
Click the Extract button and select Extract region alignment. You can then re-align as before but be sure to select Automatically determine sequences direction before clicking OK. Geneious will reverse and complement the second sequence selected before doing the alignment. Compare the dotplot for the extracted region and the re-aligned version. Notice how the diagonal line in the dotplot is now reversed.
Familiarise yourself with the display options that Geneious provides for DNA alignment. Try experimenting with alignment view settings as you will get some interesting effects.
As we have discovered here, the defaults are not necessarily the best for every situation and those that we have used here are also not ideal in every case. When you are ready, move on to the final exercise.