tag:blogger.com,1999:blog-8499895524521663926.post5862724245702230159..comments2024-03-27T07:13:39.236-04:00Comments on Phylogenetic Tools for Comparative Biology: New test "full" version of make.simmapLiam Revellhttp://www.blogger.com/profile/04314686830842384151noreply@blogger.comBlogger13125tag:blogger.com,1999:blog-8499895524521663926.post-15333517043868549622017-07-24T14:54:41.189-04:002017-07-24T14:54:41.189-04:00Hi Liam,
Great function!
I was wondering if I cou...Hi Liam, <br />Great function!<br />I was wondering if I could fix some ancestral states in this function.<br />Is that possible?<br />ThanksBruno Vilelahttps://www.blogger.com/profile/07959665828346852858noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-80649143677850036452016-12-25T13:59:02.626-05:002016-12-25T13:59:02.626-05:00Hi Marcin. You could instead increase alpha. If yo...Hi Marcin. You could instead increase alpha. If you set one, but not the other, the value of the second will be set automatically to maintain a constant mean of the distribution at the empirical (ML) value of the transition rate, q. That way changing only alpha or beta (but not both) changes only the shape of the gamma distribution, but not its mean.<br /><br />You might also try:<br /><br />plot(dgamma(x,alpha,beta))<br /><br />(in which x is a vector in the horizontal dimension) to see how different distributions look.Liam Revellhttps://www.blogger.com/profile/04314686830842384151noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-21054803304817748912016-11-30T09:40:15.033-05:002016-11-30T09:40:15.033-05:00Hi,
I have a question regarding gamma prior. I...Hi,<br /><br />I have a question regarding gamma prior. I'm not sure I understand it correctly. If I use default prior the mean and variance of gamma distribution is equal 1 so the shape of distribution is more similar to exponential function. The problem is what to do to change it to be more similar to log-normal. Should I just take larger value of beta like 3 or 4?<br /><br />Best<br />Marcin<br />MaPhttps://www.blogger.com/profile/06269199215406608318noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-50351797726822488842016-10-04T23:20:23.231-04:002016-10-04T23:20:23.231-04:00Hi Liam
Have you published about this?Hi Liam <br /><br />Have you published about this?cbuddenhagenhttps://www.blogger.com/profile/12740386663793698087noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-1113328353280085342014-04-02T13:12:21.851-04:002014-04-02T13:12:21.851-04:00Hi Liam,
really great stuff. Being able to underta...Hi Liam,<br />really great stuff. Being able to undertake posterior mapping in R is great. I thought that in response to your comment about priors "but this will probably not be a good prior density for your dataset" readers would be interested in the following article where we explore the effects of the gamma prior on the results (and potential provide a solution).<br />Maybe you could provide a link within the above text.<br />Best regards<br />Thomas <br /><br />http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0010473#pone-0010473-g006Thomas Couvreurhttps://www.blogger.com/profile/02208970133765262665noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-90953251595504826552013-05-01T11:50:46.368-04:002013-05-01T11:50:46.368-04:00Sorry, my mistake. It worked fine.
Great post at t...Sorry, my mistake. It worked fine.<br />Great post at the link!<br />ThanksAndréshttps://www.blogger.com/profile/04763931890352626135noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-59860260527202913962013-04-30T23:52:31.910-04:002013-04-30T23:52:31.910-04:00Probably should be "piecol" not "pi...Probably should be "piecol" not "piecolors" - that might be your problem. Sorry about that.<br /><br />See more about this <a href="http://blog.phytools.org/2013/04/showing-posterior-probabilities-from.html" rel="nofollow">here</a>.Liam Revellhttps://www.blogger.com/profile/04314686830842384151noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-74124531725736047572013-04-30T23:30:24.591-04:002013-04-30T23:30:24.591-04:00to.matrix is in the NAMESPACE of phytools now - bu...to.matrix is in the NAMESPACE of phytools now - but it may not have been in earlier version. You can use this:<br /><br />to.matrix<-function(x,seq) <br />sapply(seq,function(x,y) 1*(x==y),y=x)Liam Revellhttps://www.blogger.com/profile/04314686830842384151noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-7883912131996952862013-04-30T21:44:09.502-04:002013-04-30T21:44:09.502-04:00Sorry this was the previous post, now fixeds since...Sorry this was the previous post, now fixeds since I couldn't edit it.<br /><br />Worked wonderfully for the size, now trying to handle the colors for 3 traits I obtained the third one in green anyway.<br /><br />I tried:<br /><br />nodelabels(pie=XX$ace,piecolors=c("blue","red", "orange"),<br />cex=0.2)<br />tiplabels(pie=to.matrix (y, seq=colnames(XX$ace)),<br />piecolors=c("blue","red", "orange"),cex=0.5)<br /><br />Should I define something else?Andréshttps://www.blogger.com/profile/04763931890352626135noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-61267768000489128792013-04-30T21:38:14.801-04:002013-04-30T21:38:14.801-04:00Sorry for the double post, also in the second line...Sorry for the double post, also in the second line I got <br /><br />Error in BOTHlabels(text, tip, XX, YY, adj, frame, pch, thermo, pie, piecol, : <br /> could not find function "to.matrix"<br /><br />Which matrix is being referred?Andréshttps://www.blogger.com/profile/04763931890352626135noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-87029269231309611262013-04-30T21:19:35.093-04:002013-04-30T21:19:35.093-04:00This comment has been removed by the author.Andréshttps://www.blogger.com/profile/04763931890352626135noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-88867028601341564982013-04-30T20:33:59.844-04:002013-04-30T20:33:59.844-04:00Try this (tree is your phylogeny, mtrees is your s...Try this (tree is your phylogeny, mtrees is your set of stochastic mapped trees):<br /><br />XX<-describe.simmap(mtrees)<br />hh<-max(nodeHeights(tree))<br />plot(tree,no.margin=TRUE,label.offset=0.02*hh)<br />nodelabels(pie=XX$ace,piecolors=c("blue","red"),<br /> cex=0.5)<br />tiplabels(pie=to.matrix(y,seq=colnames(XX$ace)),<br /> piecolors=c("blue","red"),cex=0.5)<br /><br />You can adjust cex, label.offset, & piecolors as desired.<br /><br />Let me know if this works. - LiamLiam Revellhttps://www.blogger.com/profile/04314686830842384151noreply@blogger.comtag:blogger.com,1999:blog-8499895524521663926.post-13212446119347591372013-04-30T20:21:53.891-04:002013-04-30T20:21:53.891-04:00Hi Liam, first congrats on the new paper on "...Hi Liam, first congrats on the new paper on "Methods". }<br />I wonder how can I alter the size of the "pie" in <b>describe.simmap</b> to fit a big topology.<br />I can alter the size or scale of the whole drawing in RStudio but it would be wonderful to alter the size of the node-pies first. Is there a similar way to handle it like plotSimmap?<br />Same question for altering the color ? Would it be via a similar way plotSimmap does it ~ "leg<-c("blue","red")" ?Andréshttps://www.blogger.com/profile/04763931890352626135noreply@blogger.com