feed.xml

<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
<title type="text">Bacterial finches</title>
<generator uri="https://github.com/jekyll/jekyll">Jekyll</generator>
<link rel="self" type="application/atom+xml" href="http://sam217pa.github.io/feed.xml" />
<link rel="alternate" type="text/html" href="http://sam217pa.github.io" />
<updated>2016-02-06T16:16:14+01:00</updated>
<id>http://sam217pa.github.io/</id>
<author>
  <name>Samuel Barreto</name>
  <uri>http://sam217pa.github.io/</uri>
  <email>samuel.barreto8@gmail.com</email>
</author>


  

<entry>
  <title type="html"><![CDATA[Emacs in My Toolbox]]></title>
  <link rel="alternate" type="text/html" href="http://sam217pa.github.io/emacs/emacs-in-my-toolbox/" />
  <id>http://sam217pa.github.io/emacs/emacs-in-my-toolbox</id>
  <published>2016-02-06T15:42:58+01:00</published>
  <updated>2016-02-06T15:42:58+01:00</updated>
  <author>
    <name>Samuel Barreto</name>
    <uri>http://sam217pa.github.io</uri>
    <email>samuel.barreto8@gmail.com</email>
  </author>
  <content type="html">
    &lt;p&gt;Emacs is a powerful tool. A frequent habit of emacs user is to try to migrate
everything they know and use inside emacs. Emacs Gnus or mu4e, both emacs mail
clients, emacs org-mode, a GTD mode inside emacs—which is actually extremely
powerful—, emacs calc, a calculator inside emacs etc… are all attempts in this
line of conduct.&lt;/p&gt;

&lt;p&gt;As an emacs user myself, I also tried these stuff. And it works really well.&lt;/p&gt;

&lt;p&gt;As a matter of fact, I actually use it all day long. Let’s describe quickly how
I use it to work.&lt;/p&gt;

&lt;h1 id=&quot;emacs-matters&quot;&gt;Emacs Matters&lt;/h1&gt;

&lt;p&gt;I start emacs as I start my OS. I used to check mails with mu4e, but never
really loved the UI. I check my agenda for the day, built with org-mode. I set
up my main tasks for the day, putting some time-stamps on it.&lt;/p&gt;

&lt;h1 id=&quot;programming-and-documenting-at-the-same-time&quot;&gt;Programming and documenting. At the same time.&lt;/h1&gt;

&lt;p&gt;I then open the README.org of the project I’m currently working on, which
involves a lot of bash scripting, python and R. All these scripts are written
inside the README.org, with comments and all written in org-mode. Scripts are
written inside &lt;code&gt;src&lt;/code&gt; blocks, and tangled to a src directory containing all my
tangled files. I never ever edit a single tangled files. I always do it inside
org-mode, so that I can describe my work, what I try to do, what I already did
and what’s left to do at the end of the day. My workflow is thus entirely
managed inside emacs. I can put TODO flags in different parts of the project,
organise the src code in different header tree etc…&lt;/p&gt;

&lt;h1 id=&quot;org-mode-to-pdf&quot;&gt;Org-mode to pdf&lt;/h1&gt;

&lt;p&gt;When the time comes to write a report, I write it inside org-mode. I use zotero
to export my references to a &lt;code&gt;.bib&lt;/code&gt; file somewhere on my computer. Org-ref is a
mode to manage citations and references inside org-mode. I already wrote a 30
pages long report with figures, references, math and all inside org-mode. I
export to latex, and use &lt;code&gt;latexmk&lt;/code&gt; to export my report to pdf.&lt;/p&gt;

&lt;p&gt;Emacs is great. Let’s all use emacs. &lt;label for=&quot;724&quot; class=&quot;margin-toggle sidenote-number&quot;&gt;&lt;/label&gt;&lt;input type=&quot;checkbox&quot; id=&quot;724&quot; class=&quot;margin-toggle&quot; /&gt;&lt;span class=&quot;sidenote&quot;&gt;I actually use Spacemacs, a combination of Vim and Emacs. It’s great if you plan to stay all day at the computer : your pinky will thank you for that. It’s all the more true if you have it broken like me ! &lt;/span&gt;&lt;/p&gt;

    &lt;p&gt;&lt;a href=&quot;http://sam217pa.github.io/emacs/emacs-in-my-toolbox/&quot;&gt;Emacs in My Toolbox&lt;/a&gt; was originally published by Samuel Barreto at &lt;a href=&quot;http://sam217pa.github.io&quot;&gt;Bacterial finches&lt;/a&gt; on February 06, 2016.&lt;/p&gt;
  </content>
</entry>


  

<entry>
  <title type="html"><![CDATA[Lectures and Data Visualization]]></title>
  <link rel="alternate" type="text/html" href="http://sam217pa.github.io/opinions/lectures-and-datavis/" />
  <id>http://sam217pa.github.io/opinions/lectures-and-datavis</id>
  <published>2015-10-21T20:40:26+02:00</published>
  <updated>2015-10-21T20:40:26+02:00</updated>
  <author>
    <name>Samuel Barreto</name>
    <uri>http://sam217pa.github.io</uri>
    <email>samuel.barreto8@gmail.com</email>
  </author>
  <content type="html">
    &lt;section id=&quot;table-of-contents&quot; class=&quot;toc&quot;&gt;
  &lt;header&gt;
    &lt;h3&gt;&lt;i class=&quot;fa fa-book&quot;&gt;&lt;/i&gt; Overview&lt;/h3&gt;
  &lt;/header&gt;
&lt;div id=&quot;drawer&quot;&gt;
&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#good-lecturers-show-the-question&quot; id=&quot;markdown-toc-good-lecturers-show-the-question&quot;&gt;Good lecturers show the question&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#good-lecturers-have-a-good-style&quot; id=&quot;markdown-toc-good-lecturers-have-a-good-style&quot;&gt;Good lecturers have a good style.&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#good-lecturers-do-not-distort-the-question&quot; id=&quot;markdown-toc-good-lecturers-do-not-distort-the-question&quot;&gt;Good lecturers do not distort the question.&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#good-lecturers-synthesize&quot; id=&quot;markdown-toc-good-lecturers-synthesize&quot;&gt;Good lecturers synthesize.&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#good-lecturers-show-complexity&quot; id=&quot;markdown-toc-good-lecturers-show-complexity&quot;&gt;Good lecturers show complexity.&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#good-lecturers-show-details-and-perspective&quot; id=&quot;markdown-toc-good-lecturers-show-details-and-perspective&quot;&gt;Good lecturers show details and perspective&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#good-lecturers-serves-a-clear-purpose&quot; id=&quot;markdown-toc-good-lecturers-serves-a-clear-purpose&quot;&gt;Good lecturers serves a clear purpose.&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#conclusion&quot; id=&quot;markdown-toc-conclusion&quot;&gt;Conclusion&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

  &lt;/div&gt;
&lt;/section&gt;
&lt;!-- /#table-of-contents --&gt;

&lt;p&gt;Powerpoints are everywhere. In our world of science, Powerpoints are the &lt;em&gt;lingua
franca&lt;/em&gt; of lectures. Yet they are seldom used to their rightful potential. I
believe Powerpoints are often accused of the lecturer’s fault. I speak here with
the legitimacy of a student that spent most of the last six years listening to
lecturers with poor lecturing skills. &lt;label for=&quot;776&quot; class=&quot;margin-toggle sidenote-number&quot;&gt;&lt;/label&gt;&lt;input type=&quot;checkbox&quot; id=&quot;776&quot; class=&quot;margin-toggle&quot; /&gt;&lt;span class=&quot;sidenote&quot;&gt;I am not saying here that all of the lectures I had were of poor quality. I say most of them. &lt;/span&gt;&lt;/p&gt;

&lt;p&gt;I am a great admirer of Edward Tufte’s work about data visualization. Yet I
disagree with his opinion that Powerpoints kills reasoning. Actually I believe
there are many overlaps between good data visualization and good lectures.&lt;/p&gt;

&lt;p&gt;Here are Tufte’s opinion on graphical excellence. Good graphics :&lt;/p&gt;

&lt;blockquote&gt;
  &lt;ul&gt;
    &lt;li&gt;show the data&lt;/li&gt;
    &lt;li&gt;induce the viewer to think about substance rather than methodology, design
or something else.&lt;/li&gt;
    &lt;li&gt;avoid distorting the data.&lt;/li&gt;
    &lt;li&gt;present many number in a small space.&lt;/li&gt;
    &lt;li&gt;make large data sets coherent.&lt;/li&gt;
    &lt;li&gt;reveal the data at several levels of details, from a broad overview to a
fine structure.&lt;/li&gt;
    &lt;li&gt;serve a clear purpose.&lt;/li&gt;
    &lt;li&gt;be closely integrated with the statistical and verbal description of the
data.&lt;/li&gt;
  &lt;/ul&gt;
&lt;/blockquote&gt;

&lt;p&gt;As a matter of fact, a good lectures shares the same objectives. Good lectures :&lt;/p&gt;

&lt;ul&gt;
  &lt;li&gt;show the question.&lt;/li&gt;
  &lt;li&gt;induce the listener to think about substance rather than the lecturer’s style of lecturing.&lt;/li&gt;
  &lt;li&gt;avoid distorting the question&lt;/li&gt;
  &lt;li&gt;present many complex facts and ideas in a small space.&lt;/li&gt;
  &lt;li&gt;make complex, multivariate questions coherent.&lt;/li&gt;
  &lt;li&gt;reveals the ins and outs of the question at several levels of details, from a broad overview to the finest technical details.&lt;/li&gt;
  &lt;li&gt;serve a clear purpose.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Here I’ll expose what I think makes a good lecturer &lt;label for=&quot;134&quot; class=&quot;margin-toggle sidenote-number&quot;&gt;&lt;/label&gt;&lt;input type=&quot;checkbox&quot; id=&quot;134&quot; class=&quot;margin-toggle&quot; /&gt;&lt;span class=&quot;sidenote&quot;&gt;I’ll talk indifferently of lecturer or Powerpoint. To me, a Powerpoint is as bad as is its designer. &lt;/span&gt; . It follows Tufte’s ideas on data visualization.&lt;/p&gt;

&lt;h2 id=&quot;good-lecturers-show-the-question&quot;&gt;Good lecturers show the question&lt;/h2&gt;

&lt;p&gt;A Powerpoint and a lecturer are useless if the scientific question behind it is not understood by the audience.&lt;/p&gt;

&lt;p&gt;A good Powerpoint should always have a clear title. It describes clearly and concisely the broad topic. It serves its goal even better as a question. Questions gives direction to reasoning.&lt;/p&gt;

&lt;h2 id=&quot;good-lecturers-have-a-good-style&quot;&gt;Good lecturers have a good style.&lt;/h2&gt;

&lt;p&gt;This is my main point. Good lecturers induces the listener to think about the &lt;em&gt;question&lt;/em&gt;. The listener should not have to think about the lecture’s style. To paraphrase Pr Hervé Maisonneuve :&lt;/p&gt;

&lt;blockquote&gt;
  &lt;p&gt;Good style is absence of style.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Color that serves no goal should disappear. Forget Powerpoints ready-made templates. They have no place in Science. Superfluous words should disappear. Diagrams are clear. They take the space they need, no more.&lt;/p&gt;

&lt;p&gt;Indeed, I do agree with Edward Tufte that Powerpoints should be reserved for pictures, diagrams and charts. In Science, we can put all aspects of reasoning to good use. Some people needs a visual support to information. But the excessive use of Powerpoints has shifted this support from the lecturer to the screen. We need to realize this. The support of information is the lecturer. The audience should be &lt;em&gt;listening&lt;/em&gt;&lt;label for=&quot;536&quot; class=&quot;margin-toggle sidenote-number&quot;&gt;&lt;/label&gt;&lt;input type=&quot;checkbox&quot; id=&quot;536&quot; class=&quot;margin-toggle&quot; /&gt;&lt;span class=&quot;sidenote&quot;&gt;Listening is a skill we have to re-learn. Powerpoints have made that much damage. &lt;/span&gt;. Not staring at a screen, trying to make sense of verbless sentences.&lt;/p&gt;

&lt;p&gt;Poor lecturers tend to put too much textual information on screen. As if they were hoping that it could rescue their audience’s lack of attention…&lt;/p&gt;

&lt;p&gt;As does good charts, good lecturers are neutral enough to shift focus from form to content.&lt;/p&gt;

&lt;h2 id=&quot;good-lecturers-do-not-distort-the-question&quot;&gt;Good lecturers do not distort the question.&lt;/h2&gt;

&lt;p&gt;Good lecturers do not leave an opinionated point of view of the question in the listener’s mind. On the contrary, they should be subtle enough to put the seeds of good reasoning in it. The listener has the choice to make it flourish.&lt;/p&gt;

&lt;h2 id=&quot;good-lecturers-synthesize&quot;&gt;Good lecturers synthesize.&lt;/h2&gt;

&lt;p&gt;Good lecturers show many elements to the listener. They give him keys to analyze those elements. Or they can lead him to get this keys by himself &lt;label for=&quot;674&quot; class=&quot;margin-toggle sidenote-number&quot;&gt;&lt;/label&gt;&lt;input type=&quot;checkbox&quot; id=&quot;674&quot; class=&quot;margin-toggle&quot; /&gt;&lt;span class=&quot;sidenote&quot;&gt;It depends on the lecturer’s style. Even though he should not have a style. Ha. &lt;/span&gt;.&lt;/p&gt;

&lt;p&gt;But they have to be synthetic. One cannot hope to give a sufficiently broad overview of a complex topic with too much divergence from the central point.&lt;/p&gt;

&lt;h2 id=&quot;good-lecturers-show-complexity&quot;&gt;Good lecturers show complexity.&lt;/h2&gt;

&lt;p&gt;Good lecturers connect facts between them. Good lecturers connect theories between them. They connect theories to facts. They connect facts to theories. They connect it to the audience’s knowledge. One can not be a good lecturer without knowing his audience.&lt;/p&gt;

&lt;p&gt;A good lecturer’s main role is to make complex questions coherent.&lt;/p&gt;

&lt;h2 id=&quot;good-lecturers-show-details-and-perspective&quot;&gt;Good lecturers show details and perspective&lt;/h2&gt;

&lt;p&gt;A good lecturer has the ability to attach the finest details to the broader tendency in which they are included. He knows the broad phenomena as well as little peculiar details.&lt;/p&gt;

&lt;h2 id=&quot;good-lecturers-serves-a-clear-purpose&quot;&gt;Good lecturers serves a clear purpose.&lt;/h2&gt;

&lt;p&gt;It could be the #1 point. Good lecturers put things in context. Period. One can never have the listener’s attention if the listener does not know why he should listen. What is the point of this ? Why are you doing this ? Why work on this ? To what point ?&lt;/p&gt;

&lt;p&gt;Such questions are so common that a lecturer tends to forget to answer it.&lt;/p&gt;

&lt;h1 id=&quot;conclusion&quot;&gt;Conclusion&lt;/h1&gt;

&lt;p&gt;Tufte says a chart’s data to ink ratio should be maximized. Chartjunk should be eliminated.&lt;/p&gt;

&lt;p&gt;I say ideas to speak ratio is the measure of a lecture’s quality. Talkjunk is needless. So many talk are crippled with useless information that it is rather the rule than the exception.&lt;/p&gt;

&lt;p&gt;Focus should be on the lecturer. Not on the screen.&lt;br /&gt;
The lecturer’s focus should be on the listener. Not on the screen.&lt;/p&gt;

&lt;p&gt;Powerpoints are a tool to assist reasoning. Not a support of information. Not a decoration. They should be neutral. They should be the hammer of the lecturer. A tool.&lt;/p&gt;

    &lt;p&gt;&lt;a href=&quot;http://sam217pa.github.io/opinions/lectures-and-datavis/&quot;&gt;Lectures and Data Visualization&lt;/a&gt; was originally published by Samuel Barreto at &lt;a href=&quot;http://sam217pa.github.io&quot;&gt;Bacterial finches&lt;/a&gt; on October 21, 2015.&lt;/p&gt;
  </content>
</entry>


  

<entry>
  <title type="html"><![CDATA[Embed D3 Code into a markdown Jekyll Document]]></title>
  <link rel="alternate" type="text/html" href="http://sam217pa.github.io/datavis/embed-d3-code/" />
  <id>http://sam217pa.github.io/datavis/embed-d3-code</id>
  <published>2015-10-20T00:33:53+02:00</published>
  <updated>2015-10-20T00:33:53+02:00</updated>
  <author>
    <name>Samuel Barreto</name>
    <uri>http://sam217pa.github.io</uri>
    <email>samuel.barreto8@gmail.com</email>
  </author>
  <content type="html">
    &lt;p&gt;I deserve no credits for this. It is basically a copy and paste from Nick Such’s
previous &lt;a href=&quot;http://www.nicksuch.com/2014/03/26/d3-sample/&quot;&gt;post&lt;/a&gt;. It follows his
answer to a question on
&lt;a href=&quot;http://stackoverflow.com/questions/22651346/how-to-embed-a-d3-js-example-to-the-jekyll-blog-post&quot;&gt;StackOverflow&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;See
&lt;a href=&quot;https://raw.githubusercontent.com/nicksuch/nicksuch.github.io/master/_posts/2014-03-26-d3-sample.md&quot;&gt;here&lt;/a&gt;
for details about how to do it.&lt;/p&gt;

&lt;style&gt;

div.example {
  font-family: &quot;Helvetica Neue&quot;, Helvetica, Arial, sans-serif;
}

.box {
  font: 10px sans-serif;
}

.box line,
.box rect,
.box circle {
  fill: #008080;
  stroke: #000;
  stroke-width: 1.5px;
}

.box .center {
  stroke-dasharray: 3,3;
}

.box .outlier {
  fill: none;
  stroke: #ccc;
}

&lt;/style&gt;

&lt;script src=&quot;http://d3js.org/d3.v3.min.js&quot;&gt;&lt;/script&gt;

&lt;script src=&quot;http://bl.ocks.org/mbostock/raw/4061502/0a200ddf998aa75dfdb1ff32e16b680a15e5cb01/box.js&quot;&gt;&lt;/script&gt;

&lt;script&gt;

var margin = {top: 10, right: 50, bottom: 20, left: 50},
    width = 120 - margin.left - margin.right,
    height = 500 - margin.top - margin.bottom;

var min = Infinity,
    max = -Infinity;

var chart = d3.box()
    .whiskers(iqr(1.5))
    .width(width)
    .height(height);

d3.csv(&quot;/dataset/morley.csv&quot;, function(error, csv) {
  var data = [];

  csv.forEach(function(x) {
    var e = Math.floor(x.Expt - 1),
        r = Math.floor(x.Run - 1),
        s = Math.floor(x.Speed),
        d = data[e];
    if (!d) d = data[e] = [s];
    else d.push(s);
    if (s &gt; max) max = s;
    if (s &lt; min) min = s;
  });

  chart.domain([min, max]);

  var svg = d3.select(&quot;div#example&quot;).selectAll(&quot;svg&quot;)
      .data(data)
    .enter().append(&quot;svg&quot;)
      .attr(&quot;class&quot;, &quot;box&quot;)
      .attr(&quot;width&quot;, width + margin.left + margin.right)
      .attr(&quot;height&quot;, height + margin.bottom + margin.top)
    .append(&quot;g&quot;)
      .attr(&quot;transform&quot;, &quot;translate(&quot; + margin.left + &quot;,&quot; + margin.top + &quot;)&quot;)
      .call(chart);

  setInterval(function() {
    svg.datum(randomize).call(chart.duration(1000));
  }, 2000);
});

function randomize(d) {
  if (!d.randomizer) d.randomizer = randomizer(d);
  return d.map(d.randomizer);
}

function randomizer(d) {
  var k = d3.max(d) * .02;
  return function(d) {
    return Math.max(min, Math.min(max, d + k * (Math.random() - .5)));
  };
}

// Returns a function to compute the interquartile range.
function iqr(k) {
  return function(d, i) {
    var q1 = d.quartiles[0],
        q3 = d.quartiles[2],
        iqr = (q3 - q1) * k,
        i = -1,
        j = d.length;
    while (d[++i] &lt; q1 - iqr);
    while (d[--j] &gt; q3 + iqr);
    return [i, j];
  };
}

&lt;/script&gt;

&lt;div id=&quot;example&quot;&gt;&lt;/div&gt;

    &lt;p&gt;&lt;a href=&quot;http://sam217pa.github.io/datavis/embed-d3-code/&quot;&gt;Embed D3 Code into a markdown Jekyll Document&lt;/a&gt; was originally published by Samuel Barreto at &lt;a href=&quot;http://sam217pa.github.io&quot;&gt;Bacterial finches&lt;/a&gt; on October 20, 2015.&lt;/p&gt;
  </content>
</entry>


  

<entry>
  <title type="html"><![CDATA[Biased Gene Conversion]]></title>
  <link rel="alternate" type="text/html" href="http://sam217pa.github.io/evolution/biased-gene-conversion/" />
  <id>http://sam217pa.github.io/evolution/biased-gene-conversion</id>
  <published>2015-10-17T15:17:18+02:00</published>
  <updated>2015-10-17T15:17:18+02:00</updated>
  <author>
    <name>Samuel Barreto</name>
    <uri>http://sam217pa.github.io</uri>
    <email>samuel.barreto8@gmail.com</email>
  </author>
  <content type="html">
    &lt;section id=&quot;table-of-contents&quot; class=&quot;toc&quot;&gt;
  &lt;header&gt;
    &lt;h3&gt;&lt;i class=&quot;fa fa-book&quot;&gt;&lt;/i&gt; Overview&lt;/h3&gt;
  &lt;/header&gt;
&lt;div id=&quot;drawer&quot;&gt;
&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#natural-selection-hypothesis&quot; id=&quot;markdown-toc-natural-selection-hypothesis&quot;&gt;Natural Selection hypothesis&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#gene-conversion&quot; id=&quot;markdown-toc-gene-conversion&quot;&gt;Gene conversion&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#biased-gene-conversion&quot; id=&quot;markdown-toc-biased-gene-conversion&quot;&gt;Biased Gene Conversion&lt;/a&gt;    &lt;ul&gt;
      &lt;li&gt;&lt;a href=&quot;#initiation-bias&quot; id=&quot;markdown-toc-initiation-bias&quot;&gt;Initiation bias&lt;/a&gt;&lt;/li&gt;
      &lt;li&gt;&lt;a href=&quot;#gc-biased-gene-conversion&quot; id=&quot;markdown-toc-gc-biased-gene-conversion&quot;&gt;GC-biased gene conversion&lt;/a&gt;&lt;/li&gt;
    &lt;/ul&gt;
  &lt;/li&gt;
&lt;/ul&gt;

  &lt;/div&gt;
&lt;/section&gt;
&lt;!-- /#table-of-contents --&gt;

&lt;p&gt;Base composition of genomes is affected by many major process, being either
neutral or selective. The neutral model describes mutations as following the
genetic drift, which depends on population effective size. It does not represent
the genome reality very well, but it is a useful model as a null hypothesis. One
must reject the neutral hypothesis before affirming that a genomic trait is under
selection constraints.&lt;/p&gt;

&lt;h1 id=&quot;natural-selection-hypothesis&quot;&gt;Natural Selection hypothesis&lt;/h1&gt;

&lt;blockquote&gt;
  &lt;p&gt;A curious aspect of the theory of evolution is that everybody thinks he
understands it. &lt;em&gt;Jacques Monod&lt;/em&gt;&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;The natural selection hypothesis discriminate three cases of a genomic trait
being under selective constraints. The positive or directional selection leads
to the gradual erasing of variation on a given position of the genome, if this
position has a positive impact on the fitness. The purifying, or negative
selection, act against all variation, nearly all mutation are deleterious.
Typically, a mutation on conserved region of 16S rDNA in Bacteria are
deleterious : they does not spread in the population.&lt;/p&gt;

&lt;p&gt;Recently, a new process that can confound selection tracks with a neutral
process has been discovered : the biased gene conversion.&lt;/p&gt;

&lt;h1 id=&quot;gene-conversion&quot;&gt;Gene conversion&lt;/h1&gt;

&lt;p&gt;A gene conversion event takes place when the resolution of the intermediate
poroduct of homolog recombination leads to the
uni-directional—non-reciprocal—exchange of genetic information from one donor to
a receptor sequence. It is a key process of the first meiosis division, an
obligate step in eukaryotes gametogenesis.&lt;/p&gt;

&lt;p&gt;Nevertheless, if one allele has a greater chance to be the donor, the process is
biased.&lt;/p&gt;

&lt;h1 id=&quot;biased-gene-conversion&quot;&gt;Biased Gene Conversion&lt;/h1&gt;

&lt;p&gt;There are two main scenarios to explain a biased gene conversion event.&lt;/p&gt;

&lt;h2 id=&quot;initiation-bias&quot;&gt;Initiation bias&lt;/h2&gt;

&lt;p&gt;The initiation bias appears when a region on one double stranded DNA is more
often victim of double strand breaks than its homolog region on the sister
chromosome. This region is thus called a recombination hotspots, whereas its
homolog on the sister chromosome is called a recombination coldspots.
Paradoxically, on the lifespan of the hotspot, the coldspot will more often be
the donor of gene conversion events, leading to the gradual death of the
corresponding hotspot. There is an initiation bias on the gene conversion event.&lt;/p&gt;

&lt;h2 id=&quot;gc-biased-gene-conversion&quot;&gt;GC-biased gene conversion&lt;/h2&gt;

&lt;p&gt;The other bias is called GC-bias gene conversion. Indeed, it has been shown in
many eukaryotes, including yeast and human, that the G or C alleles of a gene
are statistically significantly more often the donor of a gene conversion event
than its A or T corresponding allele.&lt;/p&gt;

&lt;blockquote&gt;
  &lt;p&gt;It looks exactly as directed selection, but is &lt;em&gt;not&lt;/em&gt; an adaptative process.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;One can find traces of GC-bias gene conversion in a region if all positions are
affected, being either neutral or &lt;em&gt;even deleterious&lt;/em&gt;. Typically, gBGC is
associated with regions of high rate of recombination, actual or past. While
selection only act on non neutral regions&lt;sup id=&quot;fnref:3&quot;&gt;&lt;a href=&quot;#fn:3&quot; class=&quot;footnote&quot;&gt;1&lt;/a&gt;&lt;/sup&gt;, gBGC also acts on nearby introns
and non-functional sequences. Moreover, selection can act on a very large
genomic scale, when linkage is taken into account.&lt;/p&gt;

&lt;hr /&gt;

&lt;p&gt;gBGC has been shown to be responsible for the large scale variations of
GC-content, the so called genomic isochores; it can lead to the fixation of
deleterious mutations; and it can confound traces of selection.&lt;/p&gt;

&lt;div class=&quot;footnotes&quot;&gt;
  &lt;ol&gt;
    &lt;li id=&quot;fn:3&quot;&gt;
      &lt;p&gt;exons or regulatory sequences &lt;a href=&quot;#fnref:3&quot; class=&quot;reversefootnote&quot;&gt;&amp;#8617;&lt;/a&gt;&lt;/p&gt;
    &lt;/li&gt;
  &lt;/ol&gt;
&lt;/div&gt;

    &lt;p&gt;&lt;a href=&quot;http://sam217pa.github.io/evolution/biased-gene-conversion/&quot;&gt;Biased Gene Conversion&lt;/a&gt; was originally published by Samuel Barreto at &lt;a href=&quot;http://sam217pa.github.io&quot;&gt;Bacterial finches&lt;/a&gt; on October 17, 2015.&lt;/p&gt;
  </content>
</entry>


  

<entry>
  <title type="html"><![CDATA[Hill-Robertson Effects]]></title>
  <link rel="alternate" type="text/html" href="http://sam217pa.github.io/evolution/hill-robertson-effects/" />
  <id>http://sam217pa.github.io/evolution/hill-robertson-effects</id>
  <published>2015-10-15T19:06:40+02:00</published>
  <updated>2015-10-15T19:06:40+02:00</updated>
  <author>
    <name>Samuel Barreto</name>
    <uri>http://sam217pa.github.io</uri>
    <email>samuel.barreto8@gmail.com</email>
  </author>
  <content type="html">
    &lt;section id=&quot;table-of-contents&quot; class=&quot;toc&quot;&gt;
  &lt;header&gt;
    &lt;h3&gt;&lt;i class=&quot;fa fa-book&quot;&gt;&lt;/i&gt; Overview&lt;/h3&gt;
  &lt;/header&gt;
&lt;div id=&quot;drawer&quot;&gt;
&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#selective-sweep&quot; id=&quot;markdown-toc-selective-sweep&quot;&gt;Selective Sweep&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#background-selection&quot; id=&quot;markdown-toc-background-selection&quot;&gt;Background selection&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#mullers-ratchet&quot; id=&quot;markdown-toc-mullers-ratchet&quot;&gt;Muller’s Ratchet&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#conclusion&quot; id=&quot;markdown-toc-conclusion&quot;&gt;Conclusion&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

  &lt;/div&gt;
&lt;/section&gt;
&lt;!-- /#table-of-contents --&gt;

&lt;blockquote&gt;
  &lt;p&gt;Hill-Robertson effects are interference between two locus under selection. When
recombination is weak, selection on the two locus interfere. Selection is less
efficient than when it acts on two independent locus.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;Consider &lt;em&gt;A&lt;/em&gt; and &lt;em&gt;B&lt;/em&gt;, two moderately advantageous allele, in different
population. They are under positive or directional selection. If recombination
is weak, an optimal &lt;em&gt;AB&lt;/em&gt; combination can never appear. &lt;em&gt;A&lt;/em&gt; and &lt;em&gt;B&lt;/em&gt; interfere
with each other.&lt;/p&gt;

&lt;p&gt;Those are the typical Hill-Robertson effects, as described by Hill and Robertson
in 1966. They have been described under other declination since.&lt;/p&gt;

&lt;h2 id=&quot;selective-sweep&quot;&gt;Selective Sweep&lt;/h2&gt;

&lt;p&gt;Consider &lt;em&gt;A&lt;/em&gt;, a strongly advantageous allele, and &lt;em&gt;b&lt;/em&gt; a weakly deleterious
allele. &lt;em&gt;b&lt;/em&gt; is linked to &lt;em&gt;A&lt;/em&gt;&lt;label for=&quot;189&quot; class=&quot;margin-toggle sidenote-number&quot;&gt;&lt;/label&gt;&lt;input type=&quot;checkbox&quot; id=&quot;189&quot; class=&quot;margin-toggle&quot; /&gt;&lt;span class=&quot;sidenote&quot;&gt;They are in &lt;em&gt;linkage desequilibrium&lt;/em&gt; &lt;/span&gt;. The advantageous effects of &lt;em&gt;A&lt;/em&gt; leads to its
invasion in the population. It can even lead to a point of fixation. It is thus
a selective sweep of &lt;em&gt;A&lt;/em&gt;.&lt;/p&gt;

&lt;p&gt;But &lt;em&gt;b&lt;/em&gt; is linked to &lt;em&gt;A&lt;/em&gt;. &lt;em&gt;b&lt;/em&gt; will also invade the population, even if it is
deleterious. This is a case of “genetic hitch-hiking”&lt;label for=&quot;751&quot; class=&quot;margin-toggle sidenote-number&quot;&gt;&lt;/label&gt;&lt;input type=&quot;checkbox&quot; id=&quot;751&quot; class=&quot;margin-toggle&quot; /&gt;&lt;span class=&quot;sidenote&quot;&gt;As coined by John Maynard Smith &lt;/span&gt;.&lt;/p&gt;

&lt;h2 id=&quot;background-selection&quot;&gt;Background selection&lt;/h2&gt;

&lt;p&gt;It is the opposite case of selective sweep interference. Consider now &lt;em&gt;B&lt;/em&gt;, a
strongly deleterious allele, and &lt;em&gt;a&lt;/em&gt; a weakly advantageous allele. &lt;em&gt;a&lt;/em&gt; is linked
to &lt;em&gt;B&lt;/em&gt;. The deleterious effects of &lt;em&gt;B&lt;/em&gt; lead to the extinction of &lt;em&gt;B&lt;/em&gt; carrying
individuals.&lt;/p&gt;

&lt;p&gt;But &lt;em&gt;a&lt;/em&gt; is linked to &lt;em&gt;B&lt;/em&gt;. &lt;em&gt;a&lt;/em&gt; will disappear from the population, even if it is
advantageous. This is a case of background selection. All polymorphism linked to
a strongly deleterious allele is purged.&lt;/p&gt;

&lt;h2 id=&quot;mullers-ratchet&quot;&gt;Muller’s Ratchet&lt;/h2&gt;

&lt;blockquote&gt;
  &lt;p&gt;Muller’s ratchet was first described in 1932.&lt;/p&gt;
&lt;/blockquote&gt;

&lt;p&gt;It happens in clonal species, with small population sizes. A population carrying
no deleterious mutation is a population subgroup. When the first deleterious
mutation appears, the population ratchets up on the mutational burden. Since
recombination is weak, given the small population size, this mutation has no
chance to be cured. When further deleterious mutation occurs, population
degenerate. Its fitness irresistibly drops down, to the point of extinction.&lt;/p&gt;

&lt;h2 id=&quot;conclusion&quot;&gt;Conclusion&lt;/h2&gt;

&lt;p&gt;Hill-Robertson effects only occurs when recombination is weaker than mutation.
Polymorphism decreases and selection efficacy decreases too. One must consider
selection intensity, mutation and recombination respective rates. The size of
the genomic window affected depends on those three parameters : if recombination
is weak and selection strong, genome can be affected on a large scale.&lt;/p&gt;

&lt;p&gt;Hill-Robertson effects can explain counter intuitive observations. The spread of
deleterious allele can be explained by genomic interference, as is the case with
cystic fibrosis.&lt;/p&gt;

    &lt;p&gt;&lt;a href=&quot;http://sam217pa.github.io/evolution/hill-robertson-effects/&quot;&gt;Hill-Robertson Effects&lt;/a&gt; was originally published by Samuel Barreto at &lt;a href=&quot;http://sam217pa.github.io&quot;&gt;Bacterial finches&lt;/a&gt; on October 15, 2015.&lt;/p&gt;
  </content>
</entry>


  

<entry>
  <title type="html"><![CDATA[Starting a blog]]></title>
  <link rel="alternate" type="text/html" href="http://sam217pa.github.io/test-post/" />
  <id>http://sam217pa.github.io/test-post</id>
  <published>2015-10-14T16:38:14+02:00</published>
  <updated>2015-10-14T16:38:14+02:00</updated>
  <author>
    <name>Samuel Barreto</name>
    <uri>http://sam217pa.github.io</uri>
    <email>samuel.barreto8@gmail.com</email>
  </author>
  <content type="html">
    &lt;section id=&quot;table-of-contents&quot; class=&quot;toc&quot;&gt;
  &lt;header&gt;
    &lt;h3&gt;&lt;i class=&quot;fa fa-book&quot;&gt;&lt;/i&gt; Overview&lt;/h3&gt;
  &lt;/header&gt;
&lt;div id=&quot;drawer&quot;&gt;
&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#goal&quot; id=&quot;markdown-toc-goal&quot;&gt;Goal&lt;/a&gt;&lt;/li&gt;
  &lt;li&gt;&lt;a href=&quot;#an-aspiring-bacterial-evolutionnist&quot; id=&quot;markdown-toc-an-aspiring-bacterial-evolutionnist&quot;&gt;An aspiring bacterial evolutionnist&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

  &lt;/div&gt;
&lt;/section&gt;
&lt;!-- /#table-of-contents --&gt;

&lt;p&gt;Following the advice of 
&lt;a href=&quot;http://www.amazon.com/Pragmatic-Thinking-Learning-Refactor-Programmers/dp/1934356050&quot;&gt;Andy Hunt&lt;/a&gt;,
I decided to start a blog. I want to be able to communicate my ideas in the most
limpid way. Being a non-native English speaker, it is crucial for me to get used
to write in English.&lt;/p&gt;

&lt;h1 id=&quot;goal&quot;&gt;Goal&lt;/h1&gt;

&lt;p&gt;I am currently learning population genomics, applied to microbial communities.
Me and my supervisors are studying a biased gene conversion process in bacteria,
using a natural transformation model of recombination : &lt;em&gt;Acinetobacter&lt;/em&gt;.&lt;/p&gt;

&lt;p&gt;I have never been exposed to such concepts, yet they seem really appealing to
me. I want to be able to put some clear reasoning on concepts I have learned so
far. Writing a blog is perfect for that.&lt;/p&gt;

&lt;h1 id=&quot;an-aspiring-bacterial-evolutionnist&quot;&gt;An aspiring bacterial evolutionnist&lt;/h1&gt;

&lt;p&gt;Here I’ll try to expose stuff that I learned during my late readings and
research paper gambling…&lt;/p&gt;

&lt;p&gt;I mostly do microbiology and evolution on a computer, using &lt;code&gt;R&lt;/code&gt; and &lt;code&gt;Python&lt;/code&gt;,
and I try my best to be good at it. I’ll certainly blog on it sometimes.&lt;/p&gt;

&lt;p&gt;I also loves &lt;a href=&quot;https://github.com/syl20bnr/spacemacs&quot;&gt;Emacs&lt;/a&gt;, on which I’ll try
to blog some other times.&lt;/p&gt;


    &lt;p&gt;&lt;a href=&quot;http://sam217pa.github.io/test-post/&quot;&gt;Starting a blog&lt;/a&gt; was originally published by Samuel Barreto at &lt;a href=&quot;http://sam217pa.github.io&quot;&gt;Bacterial finches&lt;/a&gt; on October 14, 2015.&lt;/p&gt;
  </content>
</entry>

</feed>