View source for BIN-ALI-BLAST

<div id="ABC">
<div style="padding:5px; border:1px solid #000000; background-color:#b3dbce; font-size:300%; font-weight:400; color: #000000; width:100%;">
BLAST heuristic sequence alignment
<div style="padding:5px; margin-top:20px; margin-bottom:10px; background-color:#b3dbce; font-size:30%; font-weight:200; color: #000000; ">
(BLAST algorithm and Web interface, interpretation of BLAST alignments)
</div>
</div>

{{Smallvspace}}


<div style="padding:5px; border:1px solid #000000; background-color:#b3dbce33; font-size:85%;">
<div style="font-size:118%;">
<b>Abstract:</b><br />
<section begin=abstract />
This unit introduces the BLAST algorithm and practices a BLAST search via the Web interface, and scripted in R.
<section end=abstract />
</div>
<!-- ============================  -->
<hr>
<table>
<tr>
<td style="padding:10px;">
<b>Objectives:</b><br />
This unit will ...
* ... introduce the BLAST algorithm;
* ... discuss interpretation of BLAST output;
* ... teach how to perform a BLAST search online, and through R scripts;
</td>
<td style="padding:10px;">
<b>Outcomes:</b><br />
After working through this unit you ...
* ... can run and interpret BLAST searches;
* ... can use BLAST online;
* ... can compose R scripts to execute BLAST searches.
</td>
</tr>
</table>
<!-- ============================  -->
<hr>
<b>Deliverables:</b><br />
<section begin=deliverables />
<li><b>Time management</b>: Before you begin, estimate how long it will take you to complete this unit. Then, record in your course journal: the number of hours you estimated, the number of hours you worked on the unit, and the amount of time that passed between start and completion of this unit.</li>
<li><b>Journal</b>: Document your progress in your [[FND-Journal|Course Journal]]. Some tasks may ask you to include specific items in your journal. Don't overlook these.</li>
<li><b>Insights</b>: If you find something particularly noteworthy about this unit, make a note in your [[ABC-Insights|'''insights!''' page]].</li>
<section end=deliverables />
<!-- ============================  -->
<hr>
<section begin=prerequisites />
<b>Prerequisites:</b><br />
This unit builds on material covered in the following prerequisite units:<br />
*[[BIN-ALI-Optimal_sequence_alignment|BIN-ALI-Optimal_sequence_alignment (Optimal global and local sequence alignment)]]
<section end=prerequisites />
<!-- ============================  -->
</div>

{{Smallvspace}}



{{Smallvspace}}


__TOC__

{{Vspace}}


=== Evaluation ===
<b>Evaluation: NA</b><br />
<div style="margin-left: 2rem;">This unit is not evaluated for course marks.</div>
== Contents ==

[http://www.ncbi.nlm.nih.gov/blast '''BLAST'''] is by a margin the most important computational tool of molecular biology. It is so important, that we couldn't even begin our explorations here without it, and thus we have already used BLAST in the [[BIN-Storing_data]] unit, to find the most similar sequence to <code>MBP1_SACCE</code> in MYSPE.


{{Task|1=
*Read the introductory notes on {{ABC-PDF|BIN-ALI-BLAST|the BLAST algorithm for fast, heuristic, database-scale sequence alignment}} to discover homologues.
}}

{{Vspace}}

A good, detailed introduction to finding homologues on a database scale - and recognizing whether similar sequences are indeed homologous - is Pearson (2013)<ref>{{#pmid:23749753}}</ref>.

{{Vspace}}

==Reciprocal Best Matches==

{{Vspace}}


<div class="colmask doublepage">
  <div class="colleft">
    <div class="col1">
      <!-- Column 1 start -->

In this unit we will use BLAST to perform '''Reciprocal Best Matches'''.

One of the important questions of ''model-organism based inference'' is: which genes perform the same function in two different organisms. In the absence of other information, our best guess is that these are the two genes that are '''mutually''' most similar. The keyword here is '''mutually'''. If <code>MBP1_SACCE</code> from ''S. cerevisiae'' is the best match to <code>RES2_SCHPO</code> in ''S. pombe'', the two proteins are only mutually most similar if  <code>RES2_SCHPO</code> is more similar to <code>MBP1_SACCE</code> than to any other ''S. cerevisiae'' protein. We call this a '''Reciprocal Best Match''', or "RBM"<ref>Note that RBMs are usually orthologues, but the definition of orthologue and RBM is not the same. Most importantly, many orthologues are not RBMs. We will explore this more when we discuss phylogenetic inference.</ref>.

The argument is summarized in the figure on the right: genes that evolve under continuos selective pressure on their function have relatively lower mutation rates and are thus more similar to each other, than genes that undergo neo- or sub-functionalization after duplication.

However, there is a catch: proteins are often composed of multiple domains that implement distinct roles of their function. Under the assumptions above we could hypothesize:
* a gene in MYSPE that has the "same" function as the Mbp1 cell-cycle checkpoint switch in yeast should be an RBM to Mbp1;
* a gene that binds to the same DNA sites as Mbp1 should have a DNA-binding domain that is an RBM to the DNA binding domain of Mbp1.

Thus we'll compare RBMs in MYSPE for full-length <code>Mbp1_SACCE</code> and its DNA-binding domain, and see if the results are the same.


      <!-- Column 1 end -->
    </div>
    <div class="col2">
      <!-- Column 2 start -->
[[Image:RBM.jpg|frame|A hypothetical phylogenetic gene tree. "S" is a speciation in the tree, "D" is a duplication within a species. The duplicated gene (teal triangle) evolves towards a different function and thus acquires more mutations than its paralogue (teal circle). If an RBM search start from the teal triangle, it finds the red circle. However the reciprocal match finds the teal circle. The red and teal circles fulfill the '''RBM''' criterion.]]

      <!-- Column 2 end -->
    </div>
  </div>
</div>


{{Vspace}}

===Full-length RBM===

{{Vspace}}

You have already performed the first half of the experiment: matching from ''S. cerevisiae'' to MYSPE. The backward match is simple.

{{task|1=
# Access [http://www.ncbi.nlm.nih.gov/blast '''BLAST'''] and follow the link to the '''protein blast''' program.
# Enter the RefSeq ID for <code>MBP1_MYSPE</code> in the '''Query sequence''' field.
# Select <code>refseq_protein</code> as the '''database''' to search in, and enter <code>Saccharomyces cerevisiae (taxid:4932)</code> to restrict the '''organism''' for which hits are reported.
# Run BLAST. Examine the results.

If your top-hit is <code>NP_010227</code>, you have confirmed the RBM between <code>Mbp1_SACCE</code> and <code>Mbp1_MYSPE</code>. If it is not, let me know. I expect this to be the same and would like to verify your results if it is not<ref>One such case we encountered involved a protein that has a corrupted annotation for the DNA binding domain. It appears to be the correct orthologue, but it only has the second highest BLAST score.</ref>.

}}

{{Vspace}}

===RBM for the DNA binding domain===


{{Vspace}}

The DNA-binding domain of  <code>Mbp1_SACCE</code> is called an '''APSES''' domain. If the RBM between ''Saccharomyces cerevisiae'' Mbp1 and MYSPE is truly an orthologue, we expect all of the protein's respective domains to have the RBM property as well. But let's not simply assume what we can easily test. We'll define the sequence of the APSES domain in MBP1_SACCE and MYSPE and see how these definitions reflect in a BLAST search.

{{Vspace}}

====Defining the range of the APSES domain annotation====

{{#lst:Reference annotation yeast Mbp1|CDD_APSES}}


{{Vspace}}

====Executing the forward search====

{{Vspace}}

{{task|1=
# Access [http://www.ncbi.nlm.nih.gov/blast '''BLAST'''] and follow the link to the '''protein blast''' program.
# '''Forward search:'''
## Paste only the APSES domain sequence for <code>MBP1_SACCE</code> in the '''Query sequence''' field (copy the sequence from above).
## Select <code>refseq_protein</code> as the '''database''' to search in, and enter the correct taxonomy ID for MYSPE in the '''Organism''' field.
## Run BLAST. Examine the results.
## If the top hit is the same protein you have already seen, oK. If it's not '''add it to your protein database in RStudio'''.

}}

With this we have confirmed the sequence with the most highly conserved APSES domain in MYSPE. Can we take the sequence for the reverse search from the alignment that BLAST returns? Actually, that is not a good idea. The BLAST alignment is not guaranteed to be optimal. We should do an optimal sequnece alignment instead. That is: we use two different tools here for two different purposes: we use BLAST to identify one protein as the most similar among many alternatives and we use optimal sequence alignment to determine the best alignment between two sequences. That best alignment is what we will annotate as the MYSPE APSES domain.

We will execute the sequence alignment and the reverse search in R.

{{Vspace}}

{{ABC-unit|BIN-ALI-BLAST.R}}

{{Vspace}}

== Further reading, links and resources ==
;Contents of the BLAST report
{{#pmid: 23609542}}
<!-- {{WWW|WWW_GMOD}} -->
<!-- <div class="reference-box">[http://www.ncbi.nlm.nih.gov]</div> -->
== Notes ==
<references />

{{Vspace}}


<div class="about">
<b>About ... </b><br />
&nbsp;<br />
<b>Author:</b><br />
:Boris Steipe <boris.steipe@utoronto.ca>
<b>Created:</b><br />
:2017-08-05
<b>Modified:</b><br />
:2020-09-24
<b>Version:</b><br />
:1.1
<b>Version history:</b><br />
*1.1 2020 Maintenance
*1.0 Live version 2017
*0.1 First stub
</div>

{{CC-BY}}

[[Category:ABC-units]]
{{UNIT}}
{{LIVE}}
</div>
<!-- [END] -->