Preparing data for cladistic analysis

Mike Dallwitz miked at spider.ento.csiro.au
Tue Apr 20 23:40:37 CEST 1999


- From: Charnie Craemer

> First project:
>
> Cladistic analyses of the type species "representing" the more or less 
> 250 genera of the mite superfamily Eriophyoidea.
>
> Plan of action:
>
> Create character set(s) and item descriptions with a combination of 
> DeltaAccess and Delta programs, apart from other actions like 
> developing keys, do a cladistic analysis of the taxa (type species) 
> with Hennig86.

It would be better to use 'typical' species (i.e. most typical of the 
genus) rather than 'type' species (a nomenclatural concept). It would be 
better still to use _all_ the species, or a sample of species from each 
genus, but this may be impractical.

> (1) Bearing in mind that I would like to eventually enter all species 
> of the Eriophyoidea (about 3400 species) in DELTA format, how should I 
> structure the files at the start? Should I develop one global 
> character set (3 major files in DELTA - with characters, items etc.) 
> with all characters of the Eriophyoidea (although all characters will 
> not be applicable to all genera for instance), or should I develop 
> seperate files (several character, item etc.) for example for each 
> genus/tribe/family group, later combining the sets in a larger set? 
> The same question for later expansion to other mite groups.

It's better to use a single data set from the start. It's fairly 
difficult to combine data sets later.

The fact that many characters will be inapplicable to some taxa doesn't 
matter. The inapplicability of characters should be specified in a 
DEPENDENT CHARACTERS directive. When you enter data in the DELTA Editor, 
the inapplicable characters are automatically skipped. They are also 
handled automatically in applications such as Key, Intkey, Dist, and 
PAUP.

> (2) In brief, how does one use/manipulate data in DELTA format for use 
> for cladistic analyses in Hennig86 - or please just suggest some 
> literature I can consult.

There is a TRANSLATE INTO HENNIG FORMAT option in Confor, and a sample 
file 'tohen' illustrating its use. See also TRANSLATE INTO NEXUS FORMAT 
and sample file 'tonex'. The characters and taxa required for a 
particular purpose can be selected by means of the INCLUDE/EXCLUDE 
CHARACTERS/ITEMS directives. Character states can be combined or (for 
numeric characters) defined by means of a KEY STATES directive.

These directives are described in the DELTA User's Guide. If you haven't 
already printed a copy of this manual, I'd suggest waiting for the next 
program release, which will be later this week (I hope!). In that 
release, the User's Guide will be in the form of a Word 97 document 
(instead of Typset files, as in earlier releases).

Mike Dallwitz
CSIRO Entomology, GPO Box 1700, Canberra ACT 2601, Australia
Phone: +61 2 6246 4075   Fax: +61 2 6246 4000
Email: md at ento.csiro.au  Internet: biodiversity.uno.edu/delta/



More information about the delta-l mailing list