removedegree2nodes! : option to keep the root (JuliaPhylo#181)

also: CITATION.bib that works with PkgCite, citation() now using it.
Co-authored-by: Claudia Solis-Lemus <[email protected]>

Author: cecileane

CITATION.bib

@@ -0,0 +1,54 @@
+% reference for the software itself
+@article{2017SolislemusBastideAne_PhyloNetworks,
+  author = {Sol{\'\i}s-Lemus, Claudia and Bastide, Paul and An{\'e}, C{\'e}cile},
+  title = {PhyloNetworks: A Package for Phylogenetic Networks},
+  journal = {Molecular Biology and Evolution},
+  year = {2017},
+  volume = {34},
+  number = {12},
+  pages = {3292-3298},
+  doi = {10.1093/molbev/msx235},
+}
+
+% paper on the SNaQ method
+@article{2016SolislemusAne_SNaQ,
+  author = {Sol{\'\i}s-Lemus, Claudia and An{\'e}, C{\'e}cile},
+  title = {Inferring Phylogenetic Networks with Maximum Pseudolikelihood under Incomplete Lineage Sorting},
+  journal = {PLOS Genetics},
+  year = {2016},
+  volume = {12},
+  number = {3},
+  pages = {1-21},
+  doi = {10.1371/journal.pgen.1005896},
+}
+% paper on phylogenetic comparative methods for continuous traits on networks
+@article{2018Bastide_continuoustraitevolution,
+  author = {Bastide, Paul and Sol{\'\i}s-Lemus, Claudia and Kriebel, Ricardo and William Sparks, K and An{\'e}, C{\'e}cile},
+  title = {Phylogenetic Comparative Methods on Phylogenetic Networks with Reticulations},
+  journal = {Systematic Biology},
+  year = {2018},
+  volume = {67},
+  number = {5},
+  pages = {800-820},
+  doi = {10.1093/sysbio/syy033},
+}
+% paper on phylogenetic comparative methods for discrete traits on networks
+@article{2020Karimi_baobab_discretetraitevolution,
+  author = {Karimi, Nisa and Grover, Corrinne E and Gallagher, Joseph P and Wendel, Jonathan F and An{\'e}, C{\'e}cile and Baum, David A},
+  title = {Reticulate Evolution Helps Explain Apparent Homoplasy in Floral Biology and Pollination in Baobabs (\textit{{Adansonia}}; {Bombacoideae}; {Malvaceae})},
+  journal = {Systematic Biology},
+  year = {2020},
+  volume = {69},
+  number = {3},
+  pages = {462-478},
+  doi = {10.1093/sysbio/syz073},
+}
+% paper on continuous traits with intraspecific variation
+@article{2022Teo_intraspeciesvariation,
+  author = {Teo, Benjamin and Rose, Jeffrey P. and Bastide, Paul and An{\'e}, C{\'e}cile},
+  title = {Accounting for intraspecific variation in continuous trait evolution on a reticulate phylogeny},
+  journal = {bioRxiv},
+  year = {2022},
+  doi = {10.1101/2022.05.12.490814},
+  publisher = {Cold Spring Harbor Laboratory},
+}

README.md

@@ -33,7 +33,7 @@ To get help, check
   for common questions. Join the group to post/email your questions,
   or to receive information on new versions, bugs fixed, etc.
 
-If you use the package, please cite
+If you use the package, please cite ([bibtex format here](CITATION.bib))
 
 - Claudia Solís-Lemus, Paul Bastide and Cécile Ané (2017).
   PhyloNetworks: a package for phylogenetic networks.
@@ -86,5 +86,5 @@ ancestral state reconstruction, rates):
 - Karimi, Grover, Gallagher, Wendel, Ané & Baum (2020). Reticulate evolution
   helps explain apparent homoplasy in floral biology and pollination in baobabs
   (*Adansonia*; Bombacoideae; Malvaceae).
-  [Systematic Biology](https://academic.oup.com/sysbio/advance-article/doi/10.1093/sysbio/syz073/5613901?guestAccessKey=a32e7dd3-27fd-4a13-b171-7ff5d6da0e01),
-  69(3):462-478. doi: 10.1093/sysbio/syz073.
+  Systematic Biology,
+  69(3):462-478. doi: [10.1093/sysbio/syz073](https://academic.oup.com/sysbio/advance-article/doi/10.1093/sysbio/syz073/5613901?guestAccessKey=a32e7dd3-27fd-4a13-b171-7ff5d6da0e01).

docs/src/index.md

@@ -19,6 +19,8 @@ and their use for trait evolution.
 
 ## References
 
+See their [bibtex format](https://github.com/crsl4/PhyloNetworks.jl/blob/master/CITATION.bib).
+
 for the package:
 - Claudia Solís-Lemus, Paul Bastide and Cécile Ané (2017).
   PhyloNetworks: a package for phylogenetic networks.

src/auxiliary.jl

@@ -1707,36 +1707,8 @@ end
 
 #------------------------------------
 function citation()
-    println("If you use PhyloNetworks, please cite:")
-    println("@article{phylonetworks,")
-    println("author = {Sol\'{i}s-Lemus, C. and Bastide, P. and An\'{e}, C.},")
-    println("year = {2017},")
-    println("title = {{PhyloNetworks: a package for phylogenetic networks}},")
-    println("journal = {Molecular Biology and Evolution},")
-    println("volume = {34},")
-    println("number = {12},")
-    println("pages = {3292--3298},")
-    println("pmid = {28961984}}")
-    println("\n")
-    println("If you use SNaQ, please also cite:")
-    println("@article{snaq,")
-    println("author = {Sol\'{i}s-Lemus, C. AND An\'{e}, C.},")
-    println("journal = {PLOS Genetics},")
-    println("title = {Inferring Phylogenetic Networks with Maximum Pseudolikelihood under Incomplete Lineage Sorting},")
-    println("year = {2016},")
-    println("volume = {12},")
-    println("pages = {1--21},")
-    println("number = {3},")
-    println("pmc = {PMC4780787}}")
-    println("\n")
-    println("If you use trait models in networks, please also cite:")
-    println("@article {traits,")
-    println("author = {Bastide, Paul and Sol\'{i}s-Lemus, Claudia and Kriebel, Ricardo and Sparks, Kenneth William and An{\'e}, C{\'e}cile},")
-    println("title = {Phylogenetic Comparative Methods on Phylogenetic Networks with Reticulations},")
-    println("journal = {Systematic Biology},")
-    println("volume = {67},")
-    println("number = {5},")
-    println("pages = {800--820},")
-    println("year = {2018},")
-    println("doi = {10.1093/sysbio/syy033}}")
+    bibfile = joinpath(@__DIR__, "..", "CITATION.bib")
+    out = readlines(bibfile)
+    println("Bibliography in bibtex format also in CITATION.bib")
+    println(join(out,'\n'))
 end

src/manipulateNet.jl

@@ -494,12 +494,13 @@ function fuseedgesat!(i::Integer, net::HybridNetwork, multgammas=false::Bool)
 end
 
 """
-    removedegree2nodes!(net::HybridNetwork)
+    removedegree2nodes!(net::HybridNetwork, keeproot=false::Bool)
 
 Delete *all* nodes of degree two in `net`, fusing the two adjacent edges
 together each time, and return the network.
-If the network has a degree-2 root, then the root is eliminated as well,
-leaving the network unrooted.
+If the network has a degree-2 root and `keeproot` is false,
+then the root is eliminated as well, leaving the network unrooted.
+If `keeproot` is true, then the root is kept even if it's of degree 2.
 
 See [`fuseedgesat!`](@ref).
 
@@ -520,13 +521,23 @@ julia> PhyloNetworks.removedegree2nodes!(net);
 julia> writeTopology(net) # even the root is gone
 "(#H2,S4,(((S1,(S2)#H1),(#H1,S3)))#H2);"
 
+julia> net = readTopology("((((C:0.9)I1:0.1)I3:0.1,((A:1.0)I2:0.4)I3:0.6):1.4,(((B:0.2)H1:0.6)I2:0.5)I3:2.1);");
+
+julia> PhyloNetworks.removedegree2nodes!(net, true);
+
+julia> writeTopology(net, round=true) # the root was kept
+"((C:1.1,A:2.0):1.4,B:3.4);"
+
 ```
 """
-function removedegree2nodes!(net::HybridNetwork)
-    ndegree2nodes = sum(length(n.edge) == 2 for n in net.node)
-    # caution: nodes and their indices in the 'current' network may change some of them are removed
-    for ni in 1:ndegree2nodes # empty if 0 degree-2 nodes
-        i = findfirst(n -> length(n.edge) == 2, net.node)
+function removedegree2nodes!(net::HybridNetwork, keeproot=false::Bool)
+    rootnode = net.node[net.root]
+    toberemoved(nn) = (keeproot ? length(nn.edge) == 2 && nn !== rootnode :
+                                  length(nn.edge) == 2 )
+    ndegree2nodes = sum(toberemoved.(net.node))
+    # caution: nodes and their indices in the 'current' network may change when some of them are removed
+    for _ in 1:ndegree2nodes # empty if 0 degree-2 nodes
+        i = findfirst(toberemoved, net.node)
         i !== nothing || error("incorrect predicted number of degree-2 nodes to remove...")
         fuseedgesat!(i, net)
     end

test/test_manipulateNet.jl

@@ -217,3 +217,11 @@ net=readTopology("(4,((1,(2)#H7:::0.864):2.069,(6,5):3.423):0.265,(3,#H7:::0.136
 @test [e.number for e in net.node[12].edge] == [1,12,9] # or: rotate didn't work at node -2
 
 end # of testset for rotate
+
+@testset "other in manipulateNet" begin
+
+net0 = readTopology("((((C:0.9)I1:0.1)I3:0.1,((A:1.0)I2:0.4)I3:0.6):1.4,(((B:0.2)H1:0.6)I2:0.5)I3:2.1);");
+PhyloNetworks.removedegree2nodes!(net0, true) # true: to keep the root of degree-2
+@test writeTopology(net0, round=true) == "((C:1.1,A:2.0):1.4,B:3.4);"
+
+end # of testset for other functions in manipulateNet