<div dir="ltr">There is also an implementation of weighted ensembles in julia, last updated 8 days ago. <div><br></div><div> <a href="https://github.com/gideonsimpson/WeightedEnsemble.jl">https://github.com/gideonsimpson/WeightedEnsemble.jl</a><br></div><div><br></div><div><a href="http://juliahub.com">juliahub.com</a> sells julia instances in the cloud, the first few are free.</div><div><br></div><div>-- rec --</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Aug 30, 2021 at 3:22 PM Roger Critchlow <<a href="mailto:rec@elf.org">rec@elf.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Not a bad rant, sounds like Nick´s emergent book club, but that is not the rant I was looking for.<div><br></div><div>Yah, found the references the hard way.<div><br></div><div>The paper from LANL was <a href="https://journals.aps.org/prb/abstract/10.1103/PhysRevB.57.R13985" target="_blank">https://journals.aps.org/prb/abstract/10.1103/PhysRevB.57.R13985</a>,</div><div>pretty bad job on producing the date and author name.</div><div><br></div><div><h3 style="box-sizing:border-box;margin:0.2rem 0px 1rem;padding:0px;color:rgb(51,51,51);font-family:"Proxima Nova","Helvetica Neue",Helvetica,Helvetica,Arial,sans-serif;font-weight:200;line-height:1"><font size="4">Parallel replica method for dynamics of infrequent events</font></h3><h5 style="box-sizing:border-box;margin:0.2rem 0px 2px;padding:0px;color:rgb(85,85,85);font-family:"Proxima Nova","Helvetica Neue",Helvetica,Helvetica,Arial,sans-serif;font-weight:400;line-height:1;font-size:1.14286rem">Arthur F. Voter</h5><h5 style="box-sizing:border-box;margin:0.2rem 0px 15px;padding:0px;color:rgb(85,85,85);font-family:"Proxima Nova","Helvetica Neue",Helvetica,Helvetica,Arial,sans-serif;font-weight:400;line-height:1;font-size:1.14286rem">Phys. Rev. B <span style="box-sizing:border-box;font-weight:700;line-height:inherit">57</span>, R13985(R) – Published 1 June 1998</h5></div><div><br></div><div>The Pande lab paper was <a href="https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.86.4983" target="_blank">https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.86.4983</a></div><div><br></div><div><h3 style="box-sizing:border-box;margin:0.2rem 0px 1rem;padding:0px;color:rgb(51,51,51);font-family:"Proxima Nova","Helvetica Neue",Helvetica,Helvetica,Arial,sans-serif;font-weight:200;line-height:1"><font size="4">Mathematical Analysis of Coupled Parallel Simulations</font></h3><h5 style="box-sizing:border-box;margin:0.2rem 0px 2px;padding:0px;color:rgb(85,85,85);font-family:"Proxima Nova","Helvetica Neue",Helvetica,Helvetica,Arial,sans-serif;font-weight:400;line-height:1;font-size:1.14286rem">Michael R. Shirts and Vijay S. Pande</h5><h5 style="box-sizing:border-box;margin:0.2rem 0px 15px;padding:0px;color:rgb(85,85,85);font-family:"Proxima Nova","Helvetica Neue",Helvetica,Helvetica,Arial,sans-serif;font-weight:400;line-height:1;font-size:1.14286rem">Phys. Rev. Lett. <span style="box-sizing:border-box;font-weight:700;line-height:inherit">86</span>, 4983 – Published 28 May 2001</h5></div><div>i found the first from the second, found the second as #181/260 in the Pubmed listing for "pande vs" which the Pande lab uses as its list of publications.<br></div><div><br></div><div>I originally picked up on the topic in <a href="https://science.sciencemag.org/content/290/5498/1903" target="_blank">https://science.sciencemag.org/content/290/5498/1903</a></div><div><div id="gmail-m_-5162957409779026708gmail-sci-article-author-list" style="box-sizing:inherit;width:699px;margin-top:0px;font-family:Roboto,"Helvetica Neue",Helvetica,Arial,sans-serif"><div style="box-sizing:inherit"><div id="gmail-m_-5162957409779026708gmail-node370690" style="box-sizing:inherit"><div style="box-sizing:inherit"><h1 style="box-sizing:inherit;margin:0px 0px 15px;line-height:1.2;font-weight:400;font-family:"Roboto Condensed","Helvetica Neue",Helvetica,Arial,sans-serif"><div style="box-sizing:inherit"><font size="4">Screen Savers of the World Unite!</font></div></h1><div style="box-sizing:inherit;width:699px"><div style="box-sizing:inherit;font-weight:700;color:rgb(102,102,102);margin:0px;font-family:"Roboto Condensed","Helvetica Neue",Helvetica,Arial,sans-serif"><div style="box-sizing:inherit"><div id="gmail-m_-5162957409779026708gmail-content-block-markup" style="box-sizing:inherit"><div style="box-sizing:inherit"><ol id="gmail-m_-5162957409779026708gmail-contrib-group-1" style="box-sizing:inherit;margin:15px 0px 0px;padding:0px;list-style:none;font-size:0.875rem;white-space:nowrap;overflow:hidden;text-overflow:ellipsis"><li id="gmail-m_-5162957409779026708gmail-contrib-1" style="box-sizing:inherit;display:inline;padding:0px;margin:0px;text-align:left"><span style="box-sizing:inherit">Michael Shirts</span><a id="gmail-m_-5162957409779026708gmail-xref-aff-1-1" href="https://science.sciencemag.org/content/290/5498/1903#aff-1" style="box-sizing:inherit;background-color:transparent;color:rgb(55,88,138);text-decoration-line:none;font-size:10.5px;line-height:0;vertical-align:baseline;border:0px;font-weight:400" target="_blank"></a>, </li><li id="gmail-m_-5162957409779026708gmail-contrib-2" style="box-sizing:inherit;display:inline;padding:0px;margin:0px;text-align:left"><span style="box-sizing:inherit">Vijay S. Pande</span><a id="gmail-m_-5162957409779026708gmail-xref-aff-1-2" href="https://science.sciencemag.org/content/290/5498/1903#aff-1" style="box-sizing:inherit;background-color:transparent;color:rgb(55,88,138);text-decoration-line:none;font-size:10.5px;line-height:0;vertical-align:baseline;border:0px;font-weight:400" target="_blank">*</a></li></ol></div></div></div></div></div><p style="font-size:0.75rem;box-sizing:inherit;margin:0px;width:699px"><span style="box-sizing:inherit"><span style="box-sizing:inherit;font-weight:700;color:rgb(212,0,22)"></span> <span style="box-sizing:inherit">See all</span> authors and affiliations</span></p><div style="color:rgb(102,102,102);font-size:0.75rem;box-sizing:inherit;margin:15px 0px 0px;width:699px;line-height:1.1"><cite style="box-sizing:inherit">Science </cite> 08 Dec 2000:<br style="box-sizing:inherit">Vol. 290, Issue 5498, pp. 1903-1904<br style="box-sizing:inherit">DOI: 10.1126/science.290.5498.1903</div><div style="font-size:0.75rem;box-sizing:inherit;margin:15px 0px 0px;width:699px;line-height:1.1"><font color="#000000">So the original rant might have been on a BiosGroup mailing list or just in my head.</font></div><div style="font-size:0.75rem;box-sizing:inherit;margin:15px 0px 0px;width:699px;line-height:1.1"><font color="#000000">-- rec -</font><span style="color:rgb(102,102,102)">-</span></div></div></div></div></div></div></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Aug 30, 2021 at 2:33 PM Marcus Daniels <<a href="mailto:marcus@snoutfarm.com" target="_blank">marcus@snoutfarm.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">This week's Science has an article on predicting RNA structure using deep learning. The other approach you mention sounds like Rosetta.<br>
<br>
-----Original Message-----<br>
From: Friam <<a href="mailto:friam-bounces@redfish.com" target="_blank">friam-bounces@redfish.com</a>> On Behalf Of u?l? ?>$<br>
Sent: Monday, August 30, 2021 11:27 AM<br>
To: <a href="mailto:friam@redfish.com" target="_blank">friam@redfish.com</a><br>
Subject: Re: [FRIAM] Weighted Ensemble<br>
<br>
The below is the only thing my search turns up (via $ grep -i folding $(grep -li protein $(grep -l "From: [\"]*Roger Critchlow[\"]* <<a href="mailto:rec@elf.org" target="_blank">rec@elf.org</a>>" *))) I found nothing if I include "parallel" in the and. If you have other keywords, maybe it'll be more apparent. (Header still includes Google. So it's not clear to me when you started using GMail.)<br>
<br>
<br>
On 9/14/09 7:48 PM, Roger Critchlow wrote:<br>
> As I read it, the issue isn't whether structures and/or configurations <br>
> are/aren't important, the question is whether they operate according <br>
> to emergent or resultant rule sets.<br>
> <br>
> The Emergentists were betting heavily on the emergent rule set. They <br>
> believed that the variety of chemistry couldn't possibly be the result <br>
> of protons and electrons operating according to physics as they knew <br>
> it. They were right, it wasn't physics as they knew it, but the <br>
> answer turned out to be the result of configurational physics rather <br>
> than emergent principles of chemistry. They also bet that the variety <br>
> of biology couldn't be the result of chemical molecules operating <br>
> according to the chemistry they knew. And they were right again, it <br>
> wasn't chemistry as they knew it, but the answer turned out to be the <br>
> result of configurational chemistry rather than emergent priniciples <br>
> of biology.<br>
> <br>
> Chemistry and biology turn out to be ever more complicated <br>
> configurations of protons and electrons, with some neutron ballast, <br>
> operating according to the principles of quantum mechanics and <br>
> statistical mechanics. It's all physics, same particles, same forces, <br>
> same laws, no emergent forces. There are configuration forces, but <br>
> they're not emergent forces, they're subtle results of electrons <br>
> packing themselves into quantized energy levels in increasingly <br>
> complicated configurations of nuclei.<br>
> <br>
> The structure of DNA and the elaboration of molecular biology was the <br>
> last straw because it provided a purely physical mechanism for <br>
> inheritance.<br>
> <br>
> But you're right to see it as a bit of a conundrum. The Emergentists, <br>
> as McLaughlin summarizes them, were substantially correct:<br>
> configurations of atoms in molecules are the key to understanding <br>
> chemistry, there are all sorts of chemically distinctive things that <br>
> happen because of those configurations, none of those chemically <br>
> distinctive things are obvious when you play around with protons and <br>
> electrons in the physics lab. But it all turned out to be part of the <br>
> resultant of quantum mechanics, not emergent in the sense the <br>
> Emergentists had painted themselves into, so they were wrong in the <br>
> one sense they really cared about.<br>
<br>
<br>
<br>
<br>
On 8/30/21 10:43 AM, Roger Critchlow wrote:<br>
> This sounds like an algorithm for parallel protein folding that I <br>
> ranted about a long time ago. Start with some collection of <br>
> conformations; perform many different molecular dynamics simulations <br>
> from your starting points in parallel; continue with the most <br>
> promising subset. As molecular dynamics on proteins tends to find <br>
> lots of deadends, you can get a lot of improvement by tabu'ing the <br>
> known deadends and extending into conformations which don't double <br>
> back into visited regions. Seems I remember it went back to some monte-carlo work at LANL in the 1950's, Goodfellow?<br>
> <br>
> It also sounds a lot like Monte Carlo Tree search as used, for <br>
> instance, in AlphaGo.<br>
> <br>
> It boils down to how well you can distinguish promising and <br>
> unpromising branches.<br>
> <br>
> Whatever, it was in Friam before gmail, so I can't search for it. <br>
> There doesn't appear to be any search in the Friam archives, and the <br>
> years before<br>
> 2017 at <a href="https://redfish.com/pipermail/friam_redfish.com/" rel="noreferrer" target="_blank">https://redfish.com/pipermail/friam_redfish.com/</a> are all 404 anyway.<br>
> <br>
> -- rec --<br>
> <br>
> On Mon, Aug 30, 2021 at 12:39 PM uǝlƃ ☤>$ <<a href="mailto:gepropella@gmail.com" target="_blank">gepropella@gmail.com</a>> wrote:<br>
> <br>
>> In my ignorance, I've thought of weighted ensemble (WE) as a specific <br>
>> kind of novelty search. E.g. weighting toward trajectories that <br>
>> exhibit anomalies. Is that what you mean by it?<br>
>><br>
>> Also, for each of the 5 you're interested in, do you have convenient <br>
>> example cites for each/any of them? In particular, (2) and (3)? Or <br>
>> are these just ideas of places where you think WE should apply?<br>
>><br>
>> For my part, no. I haven't used WE in particular. I have a friend <br>
>> who's worked on identifying mechanical anomalies from audio <br>
>> (recordings of machines as they hum). He may have used it. I'll ask.<br>
>><br>
>> On 8/29/21 1:07 PM, Jon Zingale wrote:<br>
>>> I am presently working on learning weighted ensemble <<br>
>> <a href="https://arxiv.org/pdf/1906.00856.pdf" rel="noreferrer" target="_blank">https://arxiv.org/pdf/1906.00856.pdf</a>> sampling techniques and was <br>
>> curious if any here have worked with them before. The technique seems <br>
>> promising and has enjoyed quite a bit of success (even above MCMC <<br>
>> <a href="https://en.wikipedia.org/wiki/Markov_chain_Monte_Carlo" rel="noreferrer" target="_blank">https://en.wikipedia.org/wiki/Markov_chain_Monte_Carlo</a>>) in circles <br>
>> concerned with reaction rates for rare events.<br>
>>><br>
>>> Some points of interest for me include:<br>
>>><br>
>>> 1. A better sampling of fringe-outlier works/art from streaming<br>
>> services.<br>
>>> 2. An alternative (bin-based sampling) to globally defined "fitness"<br>
>> measures in evolutionary modeling.<br>
>>> 3. An application of diffusion-limited aggregation to general <br>
>>> search<br>
>> (especially in the face of limited resources)<br>
>>> 4. An application of linear logic to optimization problems in<br>
>> conformation prediction <<br>
>> <a href="https://en.wikipedia.org/wiki/Protein_structure_prediction" rel="noreferrer" target="_blank">https://en.wikipedia.org/wiki/Protein_structure_prediction</a>>.<br>
>>> 5. Investigation of dynamical properties, such as distribution of<br>
>> trajectories with "high winding number", on strange attractors.<br>
>>><br>
>>><br>
>>> While I am just beginning to grok the technique, I thought it might <br>
>>> be<br>
>> fruitful to ask here.<br>
<br>
<br>
--<br>
☤>$ uǝlƃ<br>
<br>
- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .<br>
FRIAM Applied Complexity Group listserv<br>
Zoom Fridays 9:30a-12p Mtn GMT-6 <a href="http://bit.ly/virtualfriam" rel="noreferrer" target="_blank">bit.ly/virtualfriam</a> un/subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" rel="noreferrer" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" rel="noreferrer" target="_blank">http://friam-comic.blogspot.com/</a><br>
archives: <a href="http://friam.471366.n2.nabble.com/" rel="noreferrer" target="_blank">http://friam.471366.n2.nabble.com/</a><br>
- .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .<br>
FRIAM Applied Complexity Group listserv<br>
Zoom Fridays 9:30a-12p Mtn GMT-6 <a href="http://bit.ly/virtualfriam" rel="noreferrer" target="_blank">bit.ly/virtualfriam</a><br>
un/subscribe <a href="http://redfish.com/mailman/listinfo/friam_redfish.com" rel="noreferrer" target="_blank">http://redfish.com/mailman/listinfo/friam_redfish.com</a><br>
FRIAM-COMIC <a href="http://friam-comic.blogspot.com/" rel="noreferrer" target="_blank">http://friam-comic.blogspot.com/</a><br>
archives: <a href="http://friam.471366.n2.nabble.com/" rel="noreferrer" target="_blank">http://friam.471366.n2.nabble.com/</a><br>
</blockquote></div>
</blockquote></div>