Differences with other programs
(updated February 2001)
Mesquite's unusual design
Mesquite has different restrictions and capabilities than other related programs used in phylogenetic analysis, and as such may confuse the user who assumes it will behave in a familiar way. Here are some of ways Mesquite differs from some or all available other phylogenetics programs:
- Modular — Mesquite's modularity means that its available options depend on what modules are installed. This means that menus may change as modules are installed. It also means that documentation is fluid.
- Multiple taxa blocks — Mesquite allows for multiple sets of taxa to be included in a single file. This is useful, for instance, for gene trees to be related to species trees (one set of taxa being species, the other being genes). However, it means that data files written by Mesquite might not be accepted by other programs expecting only one set of taxa.
- Multiple characters blocks — Mesquite allows for multiple data matrices of varying types to be included in a single file. Other programs put restrictions on the number, and may have trouble reading Mesquite's files.
- Tree and data matrix editors open simultaneously — Mesquite allows a tree window and a data matrix editor to be open simultaneously.
- NEXUS file format differences — Mesquite generally follows the NEXUS standard for files, but there may be the occasional difficulty transferring some NEXUS files between Mesquite and other programs. See the page on known problems.
- Scriptable with control flow — Mesquite is scriptable via its own scripting language, which allows variables and control flow (e.g., loops and if's).
Mesquite's features compared with other programs
Mesquite's capabilities are not restricted to phylogenetic biology (e.g., with population genetics and morphometrics modules in development), but since its roots are there, a comparison will be made with two phylogenetics programs: MacClade and PAUP.
Things that MacClade 4 can do that Mesquite can't (currently) do
- DNA sequence alignment — MacClade's extensive tools for manual alignment and sequence visualization are not in Mesquite.
- Other data editing features — On-the-fly protein translation and various other functions (e.g., Speak Matrix, Search & Replace, Undo) are available in MacClade's data editor but not Mesquite.
- Parsimony assumptions — MacClade has the ability to handle ordered, irreversible, Dollo, and stratigraphic characters, that Mesquite cannot.
- Summaries of parsimony reconstructions — MacClade has much more extensive and sophisticated tools for summarizing parsimony reconstructions of character evolution (Trace all changes, equivocal cycling, the Changes and Stasis chart).
- Other features — MacClade has many other features scattered around the program that aren't in Mesquite (e.g., character state tree editor, PAUP command generation, importing sequences, HTML output, saving PICT graphics files of tree images)
Things that PAUP* 4 can do that Mesquite can't (currently) do
- Fast tree searches — Mesquite can do tree searches, but they are much much slower than those available in PAUP*. However, because of the flexibility of Mesquite's criteria, there are some searches that Mesquite can do that PAUP* can't.
- Likelihood calculations — Mesquite currently has only the beginnings of likelihood calculations.
Things that Mesquite can do that MacClade and PAUP* can't (currently) do
Because Mesquite is modular, its capabilities may change rapidly.
- Flexibility of data types — Mesquite can handle continuous-valued data, and it will be easy to add new sorts of data in the future (e.g., meristic data, geographic data).
- Scriptable with flow control — Mesquite can be scripted (programmed) with loops and ifs.
- Coalescence, morphometrics, and other calculations — Mesquite has many calculations and presentation styles not available in other programs currently. Check out the page on what Mesquite can do.
© W. Maddison & D. Maddison 2001
