Feature/morphology2

PartitionFinderMorphology.py

@@ -0,0 +1,23 @@
+#Copyright (C) 2012 Robert Lanfear and Brett Calcott
+#
+#This program is free software: you can redistribute it and/or modify it
+#under the terms of the GNU General Public License as published by the
+#Free Software Foundation, either version 3 of the License, or (at your
+#option) any later version.
+#
+#This program is distributed in the hope that it will be useful, but
+#WITHOUT ANY WARRANTY; without even the implied warranty of
+#MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+#General Public License for more details. You should have received a copy
+#of the GNU General Public License along with this program.  If not, see
+#<http://www.gnu.org/licenses/>. PartitionFinder also includes the PhyML
+#program, the RAxML program, the PyParsing library, and the python-cluster library
+#all of which are protected by their own licenses and conditions, using
+#PartitionFinder implies that you agree with those licences and conditions as well.
+
+import sys
+from partfinder import main
+
+if __name__ == "__main__":
+    # Well behaved unix programs exits with 0 on success...
+    sys.exit(main.main("PartitionFinderMorphology", "morphology"))

README.md

@@ -35,4 +35,37 @@ This will run the included example analysis for PartitionFinder. More generally,
 where <PartitionFinder.py> is the full file-path to the PartitionFinder.py file
 and <foldername> is the full filepath to a folder with a phylip alignemnt and associated .cfg file.
 
+#Morphology notes.
+*Partition Finder Morphology is in beta. Tests are passing, the software runs on test datasets. But it is still unreleased software. Please see the software terms of use for further info. See below for caveats on morphology.*
+Currently, PartitionFinder works for morphology, when using the code in  branch feature/morphology2. To use this:
+
+1. Clone the repository, as above.
+2. Open Terminal (on a Mac) or Command Prompt (on Windows) and cd to the directory with PartitionFinder in it
+3. Type 'git checkout feature/morphology2'
+4. Now, PartitionFinder morphology can be executed. Similar to the nucelotide or protein versions, type:
+```python
+    python PartitionFinderMorphology.py morph/ cmdline-extras 'asc-corr=lewis'
+```    
+to execute the example, or replace <morph/> with your own folder containing a Phylip file and .cfg.
+
+*Morphology has some special caveats.*
++ If you use automated partition discovery, this is based on the [TIGER](http://bioinf.nuim.ie/tiger/) of Cummins and McInerney (2011). Because this identifies characters that are dissimilar to other characters in the matrix (see paper for a discussion), some partitions may be quite small. That a partitioning scheme is supported statistically is not a guarantee that you will estimate a more correct topology using it, or that a Bayesian topology search will arrive at convergence using this scheme. We strongly suggest comparing trees between partitioned and unpartitioned runs, and, as always, _performing multiple rounds of topology estimation for any given set of parameters._
++ If all the data in your dataset are binary, specify 'binary' on line 10.
++ The command line option 'asc-corr=lewis' is an ascertainment bias correction. Unless you have collected invarient sites, you 
+need to specify an ascertainment correction. 'asc-corr=lewis' is the correction described in Paul Lewis' [2001](http://sysbio.oxfordjournals.org/content/50/6/913) paper introducing the Mk model. More information can be found on 
+the RAxML [website](http://sco.h-its.org/exelixis/resource/download/NewManual.pdf).
++ In order to take advantage of these important corrections, make sure you are using at least version 8.1.13
+from the RAxML [github](https://github.com/stamatak/standard-RAxML/releases). 
++ Phylip is not a standard format for morphology, but it is very simple. It is simply the a one-line header with the name of species 
+and characters, separated by a space. However, you do need to remove spaces from species names. If you would rather do this programmatically,
+in the helper_scripts directory, find the script converter.py. It is called via:
+```python
+    python converter.py 'path to files' 'files extension' 'format of input files' 'format you need'
+```
+For example, to convert a nexus file in the current working directory, type:
+```python
+    python converter.py . .nex nexus phylip
+```
+This script depends on the Dendropy [library](https://pythonhosted.org/DendroPy/index.html).
+
 For more details, read the manual.

docs/downloadplot.r

@@ -2,26 +2,33 @@ library(ggplot2)
 library(reshape)
 
 dat <- read.csv("~/Documents/Projects_Current/partitionfinder/docs/downloadstats.csv")
-dat.org <- dat
-
-#the version changes in march and may need to be fixed up
-dat <- dat[,-c(4, 5, 6)]
-
-dat <- aggregate(. ~ date, data=dat, FUN=sum)
-
+dat <- aggregate(. ~ date, data=dat, FUN=sum, na.action=na.pass)
 dat <- dat[order(as.Date(dat$date, format="%d/%m/%Y")),]
 
-
 dat$Mac <- cumsum(dat$Mac)
 dat$Windows <- cumsum(dat$Windows)
 
-dat <- melt(dat)
 
-colnames(dat) <- c("date", "OS", "downloads")
+# Download plot
+dl <- melt(dat[,c(1, 2, 3)])
+colnames(dl) <- c("date", "OS", "downloads")
+dl$date <- as.Date(dl$date, format = "%d/%m/%Y")
 
-dat$date <- as.Date(dat$date, format = "%d/%m/%Y")
+quartz(width=10, height=5)
+p <- ggplot(dl, aes(date, downloads))
+p + geom_area(aes(colour = OS, fill= OS), position = 'stack')
+dev.copy2pdf(file="~/Documents/Projects_Current/partitionfinder/docs/downloadplot.pdf")
+dev.off()
 
 
-p <- ggplot(dat, aes(date, downloads))
+# Citation plot
+c <- dat[,c(1,5)]
+c$date <- as.Date(c$date, format = "%d/%m/%Y")
+c <- c[complete.cases(c),]
+
+quartz(width=10, height=5)
+p <- ggplot(c, aes(date, citations.googlescholar))
+p + geom_point() + geom_smooth(stat="identity")
+dev.copy2pdf(file="~/Documents/Projects_Current/partitionfinder/docs/citationsplot.pdf")
+dev.off()
 
-p + geom_area(aes(colour = OS, fill= OS), position = 'stack')

docs/downloadstats.csv

@@ -1 +1 @@
-date,Mac,Windows,version,citations.googlescholar,citations.added01/01/2012,30,0,0.9,10,1001/02/2012,66,0,0.9,,201/03/2012,28,0,0.9,,201/03/2012,44,0,0.9.2,,001/04/2012,39,0,0.9.2,,201/05/2012,70,48,1.0.0,,201/06/2012,35,58,1.0.0,,201/07/2012,41,54,1.0.0,,201/08/2012,107,190,1.0.1,,201/09/2012,68,95,1.0.1,,201/10/2012,62,65,1.0.1,,201/11/2012,50,76,1.0.1,,201/12/2012,62,126,1.0.1,,201/01/2013,88,1075,1.0.1,,201/02/2013,72,252,1.0.1,,201/03/2013,108,152,1.0.1,,201/04/2013,100,127,1.0.1,,201/05/2013,59,68,1.0.1,,301/05/2013,51,50,1.1.0,,001/06/2013,50,58,1.1.0,,301/06/2013,101,86,1.1.1,49,301/07/2013,96,114,1.1.1,56,7
+date,Mac,Windows,version,citations.googlescholar,citations.added.per.month01/01/2012,30,0,0.9,10,01/02/2012,66,0,0.9,NA,01/03/2012,28,0,0.9,NA,01/03/2012,44,0,0.9.2,NA,01/04/2012,39,0,0.9.2,NA,01/05/2012,70,48,1.0.0,NA,01/06/2012,35,58,1.0.0,NA,01/07/2012,41,54,1.0.0,NA,01/08/2012,107,190,1.0.1,NA,01/09/2012,68,95,1.0.1,NA,01/10/2012,62,65,1.0.1,NA,01/11/2012,50,76,1.0.1,NA,01/12/2012,62,126,1.0.1,NA,01/01/2013,88,1075,1.0.1,NA,01/02/2013,72,252,1.0.1,NA,01/03/2013,108,152,1.0.1,NA,01/04/2013,100,127,1.0.1,NA,01/05/2013,59,68,1.0.1,NA,01/05/2013,51,50,1.1.0,NA,01/06/2013,50,58,1.1.0,NA,01/06/2013,101,86,1.1.1,49,01/07/2013,96,114,1.1.1,56,701/08/2013,94,119,1.1.1,63,701/09/2013,188,104,1.1.1,77,1401/10/2013,123,108,1.1.1,92,15

examples/aminoacid/partition_finder.cfg

@@ -1,5 +1,5 @@
 ## ALIGNMENT FILE ##
-alignment = Als_etal_2004.phy;
+alignment = mat.nex;
 
 ## BRANCHLENGTHS: linked | unlinked ##
 branchlengths = linked;