Skip to content

Commit

Permalink
Update README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
ekmolloy authored May 3, 2023
1 parent 9b71ba7 commit 42f67e2
Showing 1 changed file with 9 additions and 9 deletions.
18 changes: 9 additions & 9 deletions example/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -42,32 +42,32 @@ The output tree should be
```
This tree has a Dollo score of `11618` and is topologically equivalent to the Dollo tree presented in Figure 2A by Korstian et al.

Because this data set has only 11 taxa, we can also analyze this data set using branch-and-bound. To perform this analysis, we performed this analysis using [PAUP*](https://paup.phylosolutions.com), although you might also try the[Dollop function distributed with Phylip](https://evolution.genetics.washington.edu/phylip/doc/dollop.html).
Because this data set has only 11 taxa, we can also analyze this data set using branch-and-bound. We performed this analysis using [PAUP*](https://paup.phylosolutions.com), although you might also try the [Dollop function distributed with Phylip](https://evolution.genetics.washington.edu/phylip/doc/dollop.html).

After downloading PAUP* to this directory, do the following (making sure to use the correct binary name):
After downloading PAUP* to this directory, use the following commands, making sure to use the correct binary name:
```
gunzip paup4a168_osx
chmod a+x paup4a168_osx
./paup4a168_osx -n paup-dollo-bnb-myotis-ves.nex
```
The resulting tree is topologically equivalent to the one recovered by Dollo-CDP. Note that PAUP* reports a Dollo score of 20939, which differs from our score because we only counts losses (1 -> 0) and does not count gains (0 -> 1). You can score the PAUP tree using the command:
The resulting tree is topologically equivalent to the one recovered by Dollo-CDP. Note that PAUP* reports a Dollo score of `20939`, which differs from our score because we only count losses (`1 -> 0`) and do not count gains (`0 -> 1`). You can score the PAUP tree using the command:
```
../../src/dollo-cdp-v1.0.0-osx -q paup-dollo-bnb-myotis-ves-strict.tre -i myotis_ves.nex
../src/dollo-cdp \
-q paup-dollo-bnb-myotis-ves-strict.tre \
-i myotis_ves.nex
```
The output score is `11618`.


To analyze data sets with larger numbers of taxa, you may wish to speed up Dollo-CDP by further constraining the space of allowed solutions. This can be achieved by performing heuristic search with PAUP*, and then giving the resulting trees to Dollo-CDP. Let's create a constraints by building a starting tree via random taxon addition and then performing TBR moves, saving the 100 best scoring trees.

To analyze data sets with larger numbers of taxa, you may wish to speed-up Dollo-CDP by further constraining the space of allowed solutions. This can be achieved by performing heuristic search with PAUP* and then giving the resulting trees to Dollo-CDP as constraints. Let's create a constraints by building a starting tree via random taxon addition and then performing TBR moves, saving the 100 best scoring trees.
```
./paup4a168_osx -n paup-dollo-hsearch-myotis-ves.nex
```
Now let's give these trees as constrains to Dollo-CDP.
Now let's give these trees as constraints to Dollo-CDP.
```
../src/dollo-cdp \
-i myotis_ves.nex \
-g Davi \
-t paup-dollo-hsearch-myotis-ves-all.trees \
-o dollo-cdp-plus-hsearch-constraints-myotis-ves.tre
```
This gives us the same tree as the other two analyses.
This gives us the same tree as the other two analyses with Dollo parsimony.

0 comments on commit 42f67e2

Please sign in to comment.