<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi William,<div><br></div><div>This sounds very interesting indeed!<br><div><br></div><div>It would be great to have such linear method implemented in GenA :)</div><div><br></div><div>best wishes,</div><div>Yurii<br><div>
<span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: -webkit-auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; font-size: medium; "><span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: -webkit-auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; font-size: medium; "><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: -webkit-auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; font-size: medium; "><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div><div>-------------------------------------------------------</div><div>Yurii Aulchenko, PhD, Dr. Habil.</div><div>Independent researcher and consultant</div><div>yurii [dot] aulchenko [at] gmail [dot] com</div></div></div></span></div></span></span>
</div>
<br><div><div>On Oct 11, 2011, at 9:55 PM, Astle, William J wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">
Hi all,
<div><br>
</div>
<div>This might be of interest, </div>
<div><br>
</div>
<div>best wishes </div>
<div><br>
</div>
<div>Will</div>
<div><br>
</div>
<div>></div>
<div>></div>
<div>> Dear William,
<div>></div>
<div>> My name is Christoph Lippert. I am a PhD student from the Max Planck Institute for Developmental Biology in Tuebingen, currently working on Mixed Models for GWAS.</div>
<div>></div>
<div>> Recently, we had a paper in Nature Methods, where among other results we presented a way to optimize the ratio of variance parameters for genome-wide SNPs in a mixed model with only a single cubic operation.</div>
<div>>Last week, I heard that you had a similar result in your thesis. I would be highly interested in your derivation of that result.</div>
<div>> Could you perhaps send me a copy of your thesis or a writeup of that result? This would be highly appreciated.</div>
<div>></div>
<div>> Thank you,</div>
<div>> Christoph</div>
<div>></div>
<div>> P.S.: If you are interested in our paper, here is the link:</div>
<div>> <a href="http://www.nature.com/nmeth/journal/vaop/ncurrent/abs/nmeth.1681.html">http://www.nature.com/nmeth/journal/vaop/ncurrent/abs/nmeth.1681.html</a></div>
</div>
<div><br>
</div>
<div><br>
</div>
<div>
<div>
<div>
<div>
<div>On 8 Oct 2011, at 04:37, L.C. Karssen wrote:</div>
<br class="Apple-interchange-newline">
<blockquote type="cite">
<div>Dear list,<br>
<br>
I was going through the changelog of GenABEL and saw that Gulya's<br>
poly_eigen is the default method for polygenic() since version 1.6-8. I<br>
did a quick comparison on the computation speed of the three methods<br>
(polygenic_hglm, polygenic with polylik and with poly_eigen) on our<br>
population and even though I knew what to expect from Yurii's results, I<br>
am still impressed at the speedup! <br>
<br>
I was wondering, is there a publication on this method/implementation?<br>
Has there been any comparison of poly_eigen vs. the old method (in terms<br>
of variance, lambda etc.? In February/March this year there was a<br>
discussion on this list where Yurii compared the polylik method with<br>
polygenic_hglm, but I haven't seen anything similar on the poly_eigen<br>
method.<br>
<br>
<br>
Enjoy the weekend,<br>
<br>
Lennart.<br>
<br>
<br>
On vr, 2011-05-20 at 21:16 +0200, Yurii Aulchenko wrote:<br>
<blockquote type="cite">In attached PDF I show the timing for running 'polygenic' (red line;<br>
</blockquote>
<blockquote type="cite">ML operating on Gulya's idea), polygenic_hglm (black; Xia/Lars), and<br>
</blockquote>
<blockquote type="cite">REML.rotate (green; Lars's implementation/extension of Gulya's idea)<br>
</blockquote>
<blockquote type="cite">as a function of sample size (X axes).<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">While polygenic and hglm show exponential trend, REMLrotate line is<br>
</blockquote>
<blockquote type="cite">not only the lowest, but also looks almost linear to me (also looking<br>
</blockquote>
<blockquote type="cite">at the code, it should be roughly linear on time with no of subjects).<br>
</blockquote>
<blockquote type="cite">So, with REMLrotate we can get arbitrary big speed-up (x100, x1000,<br>
</blockquote>
<blockquote type="cite">x10000, ... you name it!) cf other methods, just by adding more people<br>
</blockquote>
<blockquote type="cite">to the data :) Not a bad result!<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">Lars and Gulya, you are the champions!<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">best wishes,<br>
</blockquote>
<blockquote type="cite">Yurii<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">2011/5/19 Yurii Aulchenko <<a href="mailto:yurii.aulchenko@gmail.com">yurii.aulchenko@gmail.com</a>>:<br>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">Correction: as Lars spotted out, I did placed my 'tags' wrongly<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">concerning time performance results, so actually REML.rotate is the<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">fastest.<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">Here are updated results:<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">With conv=1e-8 mean time for estimation on 300 IDs was:<br>
</blockquote>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite"><br>
</blockquote>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">17.81 sec for polygenic_hglm<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">
<blockquote type="cite">2.27 sec for 'polygenic'-ML<br>
</blockquote>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">1.33 sec for REML.rotate<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">best wishes, and sorry for inconvenience,<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite">Yurii<br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><br>
</blockquote>
</blockquote>
<blockquote type="cite">_______________________________________________<br>
</blockquote>
<blockquote type="cite">genabel-devel mailing list<br>
</blockquote>
<blockquote type="cite"><a href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
</blockquote>
<blockquote type="cite"><a href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a><br>
</blockquote>
<br>
-- <br>
-----------------------------------------------<br>
L.C. Karssen<br>
Erasmus MC<br>
Department of Epidemiology<br>
Room Ee-2224<br>
<br>
Postbus 2040<br>
3000 CA Rotterdam<br>
The Netherlands<br>
<br>
phone: +31-10-7044217<br>
fax: +31-10-7044657<br>
e-mail: <a href="mailto:l.karssen@erasmusmc.nl">l.karssen@erasmusmc.nl</a><br>
GPG key ID: 0E1D39E3<br>
-----------------------------------------------<br>
_______________________________________________<br>
genabel-devel mailing list<br>
<a href="mailto:genabel-devel@lists.r-forge.r-project.org">genabel-devel@lists.r-forge.r-project.org</a><br>
<a href="https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel">https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/genabel-devel</a></div>
</blockquote>
</div>
<br>
</div>
</div>
</div>
</div>
</blockquote></div><br></div></div></body></html>