<div dir="ltr">Not really. In situation like this as in most cases, there is no true K - only some clustering solutions are a more efficient caricature of the data than others. <div><br></div><div>In this case, K=2, 3, ... 10 are all equivalently good caricatures.</div><div><br></div><div>Cheers</div><div>Thibaut</div></div><div class="gmail_extra"><br clear="all"><div><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div><div><div><br>--<br>Dr Thibaut Jombart<br>Lecturer, Department of Infectious Disease Epidemiology, <span style="font-size:12.8px">Imperial College London</span></div></div><div><span style="font-size:12.8px">Head of RECON: </span><span style="font-size:12.8px"><a href="http://repidemicsconsortium.org" target="_blank">repidemicsconsortium.org</a></span><br></div></div><div><a href="http://sites.google.com/site/thibautjombart/" style="font-size:12.8px" target="_blank">sites.google.com/site/thibautjombart/</a><br></div><div><a href="http://github.com/thibautjombart" target="_blank">github.com/thibautjombart</a></div>Twitter: <a href="http://twitter.com/TeebzR" target="_blank">@TeebzR</a><br></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
<br><div class="gmail_quote">On 6 December 2016 at 16:59, Felipe Hernández <span dir="ltr"><<a href="mailto:fhernandeu@uc.cl" target="_blank">fhernandeu@uc.cl</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Ok, thanks! So just putting attention in the lower k-mean value doesn't relate to the more likely number of clusters at the end? Ultimately, may K=5 be considered as the most probable number of genetic clusters explained by my dataset, or should I consider other factors too? I tried your suggestions and see what I can get. Thanks!<div><br></div><div>Best, </div></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">2016-12-06 11:13 GMT-05:00 Thibaut Jombart <span dir="ltr"><<a href="mailto:thibautjombart@gmail.com" target="_blank">thibautjombart@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hello, <div><br></div><div>the results will be a bit more stable if you increase the number of starting points for the k-means (see arg. n.start). </div><div><br></div><div>It should not really impact the outcome though: here, any K from 2 to 12 is an equally good solution, at least as judged by the BIC.</div><div><br></div><div>Cheers</div><span class="m_4265749951712451390HOEnZb"><font color="#888888"><div>Thibaut</div></font></span></div><div class="gmail_extra"><span><br clear="all"><div><div class="m_4265749951712451390m_3928019020658517623gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><div><div><div><br>--<br>Dr Thibaut Jombart<br>Lecturer, Department of Infectious Disease Epidemiology, <span style="font-size:12.8px">Imperial College London</span></div></div><div><span style="font-size:12.8px">Head of RECON: </span><span style="font-size:12.8px"><a href="http://repidemicsconsortium.org" target="_blank">repidemicsconsortium.or<wbr>g</a></span><br></div></div><div><a href="http://sites.google.com/site/thibautjombart/" style="font-size:12.8px" target="_blank">sites.google.com/site/thibautj<wbr>ombart/</a><br></div><div><a href="http://github.com/thibautjombart" target="_blank">github.com/thibautjombart</a></div>Twitter: <a href="http://twitter.com/TeebzR" target="_blank">@TeebzR</a><br></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
<br></span><div><div class="m_4265749951712451390h5"><div class="gmail_quote">On 6 December 2016 at 15:17, Felipe Hernández <span dir="ltr"><<a href="mailto:fhernandeu@uc.cl" target="_blank">fhernandeu@uc.cl</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Thanks Thibaut,<div><br></div><div>Here you have the image and values for each estimated K. Any advice is more than welcome, thanks!</div><div><br></div><div>Best,</div><div>Felipe</div><div><br></div><div><div>> grp</div><div>$Kstat</div><div> K=1 K=2 K=3 K=4 K=5 K=6 K=7 K=8 </div><div>1494.756 1481.467 1473.864 1472.002 1470.633 1472.970 1470.754 1472.011 </div><div> K=9 K=10 K=11 K=12 K=13 K=14 K=15 K=16 </div><div>1471.813 1473.632 1473.924 1476.759 1476.699 1475.433 1479.546 1481.119 </div><div> K=17 K=18 K=19 K=20 K=21 K=22 K=23 K=24 </div><div>1481.292 1485.865 1488.130 1488.356 1493.552 1494.979 1501.182 1499.258 </div><div> K=25 K=26 K=27 K=28 K=29 K=30 K=31 K=32 </div><div>1500.146 1504.113 1511.598 1511.550 1513.889 1516.275 1522.144 1524.733 </div><div> K=33 K=34 K=35 K=36 K=37 K=38 K=39 K=40 </div><div>1528.089 1530.409 1535.778 1538.049 1541.269 1546.197 1547.656 1552.127 </div><div><br></div><div>$stat</div><div> K=5 </div><div>1470.633 </div></div><div><br></div><div><br></div></div><div class="m_4265749951712451390m_3928019020658517623HOEnZb"><div class="m_4265749951712451390m_3928019020658517623h5"><div class="gmail_extra"><br><div class="gmail_quote">2016-12-05 10:10 GMT-05:00 Thibaut Jombart <span dir="ltr"><<a href="mailto:thibautjombart@gmail.com" target="_blank">thibautjombart@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear Felipe,<br>
<br>
this is always a hard question, as different methods essentially do..<br>
different things. The K-means in find.clusters optimizes the variance<br>
between groups, while BAPS maximizes a likelihood function under a<br>
given population genetics model. So it may be the case that you have<br>
~17 demes roughly at HWE, but that only 4-5 groups are optimum in<br>
terms of clearly delineated groups. And this is assuming both methods<br>
are 'right'. They may be prone to all sorts of biases. Namely, largely<br>
different group variances for the K-means, and deviations from the<br>
original model in BAPS.<br>
<br>
Feel free to post the image (or a link to it) of the BIC for<br>
find.clusters if you want a 2-cents advice on the number of K to look<br>
at.<br>
<br>
Best<br>
Thibaut<br>
<br>
--<br>
Dr Thibaut Jombart<br>
Lecturer, Department of Infectious Disease Epidemiology, Imperial College London<br>
Head of RECON: <a href="http://repidemicsconsortium.org" rel="noreferrer" target="_blank">repidemicsconsortium.org</a><br>
<a href="http://sites.google.com/site/thibautjombart/" rel="noreferrer" target="_blank">sites.google.com/site/thibautj<wbr>ombart/</a><br>
<a href="http://github.com/thibautjombart" rel="noreferrer" target="_blank">github.com/thibautjombart</a><br>
Twitter: @TeebzR<br>
<div><div class="m_4265749951712451390m_3928019020658517623m_6867035185061574274h5"><br>
<br>
On 5 December 2016 at 14:29, Felipe Hernández <<a href="mailto:fhernandeu@uc.cl" target="_blank">fhernandeu@uc.cl</a>> wrote:<br>
> Good morning,<br>
><br>
> I wonder if you may guide me with this question (that may be pretty basic<br>
> surely). After a run DaPC analysis using adegenet, I'm usually getting K<br>
> between 4 and 5 for my dataset (480 hogs, 59 microsats, 39 sampling sites).<br>
> Maximum number of clusters tried are 40. Afterwards, I tried to estimate<br>
> number of clusters (spatial clustering by individuals) using another<br>
> software (BAPS 6.0), but I got an even higher number of estimated cluster<br>
> (K=17), after testing different maximum number of K's (i.e., K=5 through<br>
> K=20). Any clue about what's the reason of this? Maybe related to the<br>
> maximum number of cluster tested? Or, linkage disequilibrium between some<br>
> loci? Sorry if the question is really basic, but I would appreciate any<br>
> advice.<br>
><br>
> Regards,<br>
> Felipe<br>
><br>
> --<br>
> Felipe Hernández<br>
> Médico Veterinario (DVM), MSc.<br>
> PhD. Candidate<br>
> Interdisciplinary Ecology Program<br>
> School of Natural Resources and Environment<br>
> Wildlife Ecology and Conservation Department<br>
> University of Florida<br>
><br>
</div></div>> ______________________________<wbr>_________________<br>
> adegenet-forum mailing list<br>
> <a href="mailto:adegenet-forum@lists.r-forge.r-project.org" target="_blank">adegenet-forum@lists.r-forge.r<wbr>-project.org</a><br>
> <a href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/adegenet-forum" rel="noreferrer" target="_blank">https://lists.r-forge.r-projec<wbr>t.org/cgi-bin/mailman/listinfo<wbr>/adegenet-forum</a><br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="m_4265749951712451390m_3928019020658517623m_6867035185061574274gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr">Felipe Hernández<div>Médico Veterinario (DVM), MSc.<div>PhD. Candidate<div>Interdisciplinary Ecology Program</div><div>School of Natural Resources and Environment</div><div>Wildlife Ecology and Conservation Department</div><div>University of Florida</div></div></div></div></div></div></div></div></div>
</div>
</div></div></blockquote></div><br></div></div></div>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="m_4265749951712451390gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr">Felipe Hernández<div>Médico Veterinario (DVM), MSc.<div>PhD. Candidate<div>Interdisciplinary Ecology Program</div><div>School of Natural Resources and Environment</div><div>Wildlife Ecology and Conservation Department</div><div>University of Florida</div></div></div></div></div></div></div></div></div>
</div>
</div></div></blockquote></div><br></div>