Dear All, I have posted the following question on evoldir: "I am a PhD student working on the hybrid complex of Pelophylax esculentus (Rana esculenta). This complex consists of diploid and triploid individuals. I am mainly working with microsatellite data and want to perform assignment analyses, genetic diversity and distances. However, most of the software currently available cannot handle triploidy. Is there someone who knows software that can perform assignment analyses with a mixed diploid/triploid dataset? And how would I best recode my data? All answers would be helpful." Because I received a lot of emails from people with the same problem, I will forward all the answers I have received. This way I hope I may help somebody else. The last email (on the bottom of this email) contained the solution for my assignment problem. Received answers: Dear Griet, I have a similar problem, but at the same time not a very good answer to it, what I recently tried was to use "popdist" http://genetics.agrsci.dk/~bernt/popgen/ with this program each allele gets its own code and it calculates genetic distances and identity. In my case I received a distance matrix which I input into QuickTree (http://mobyle.pasteur.fr/cgi-bin/MobylePortal/portal.py?form=quicktree) to get a tree in New Hampshire format which then could be transferred in any tree viewing software. I am sure this is not the most sophisticated answer and just a try, but I would be very interested in the answer you receive and to see how other people did it. Best regards Marie You might have a look at a paper by Kloda et al published in Heredity which makes use of an Fst analogue suitable for microsat data from polyploids to do a PCO type analysis. Heredity (2008) 100, 253–260; doi:10.1038/sj.hdy.6801044 Using principle component analysis to compare genetic diversity across polyploidy levels within plant complexes: an example from British Restharrows (Ononis spinosaand Ononis repens) J M Kloda1, P D G Dean2, C Maddren1, D W MacDonald1 and S Mayes3 See here: http://www.nature.com/hdy/journal/v100/n3/abs/6801044a.html Richard Nichols Professor of Genetics http://www.sbcs.qmul.ac.uk/people Hallo Griet! I would be very interested in the answers! Could you please forward to me as well??? Thank you very much!!! Marta Hi, see Meirmans, P.G., E.C. Vlot, J.C.M. Den Nijs & S.B.J. Menken: (2003), Spatial ecological and genetic structure of a mixed population of sexual diploid and apomictic triploid dandelions, Journal of Evolutionary Biology 16 p.343-352. Meirmans, P.G., & P.H. Van Tienderen: (2004), GENOTYPE and GENODIVE: two programs for the analysis of genetic diversity of asexual organisms, Molecular Ecology Notes 4 p.792-794. and http://www.bentleydrummer.nl/software/software/GenoDive.html http://www.bentleydrummer.nl/software/software/Other%20software.html -- Darren Obbard Institute of Evolutionary Biology Ashworth Labs Kings Buildings University of Edinburgh, UK darren.obbard@ed.ac.uk Lab: 0131 651 3614 -- Office [Note Change!]: 0131 650 8659 Mobile: 07968 838 635 -- Home: 0131 466 0341 http://www.biology.ed.ac.uk/research/groups/obbard/ The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. Hello Griet, I do not have the answer for your question but I'm facing the same problem with microsats in diploid/triploid apple trees. So I would be very interested in any program for analysis. Could you please forward the answers or some of them to me? thank you very much, regards Andrea Frei Molecular Diagnostics ACW Agroscope Postfach 8820 Wädenswil Switzerland I saw your query on triploid data analysis on evoldir. You should check the R package 'adegenet'. I don't know much about it, but I heard that this package deals with all levels of ploidy. http://adegenet.r-forge.r-project.org/ Don't hesitate ton contact the author: jombart@biomserv.univ-lyon1.fr I know him and I'm sure he could help you to solve this problem. Good luck, Perrine -- PELOSSE Perrine CNRS UMR 5558 - Laboratoire "Biométrie et Biologie Evolutive" Université Claude Bernard - Lyon 1 Equipe "Ecologie du Comportement et Dynamique des Populations" 43 bd du 11 novembre 1918 69622 Villeurbanne cedex - France tel : +33 (0)4 72 43 12 86 fax : +33 (0)4 72 43 13 88 pelosse@biomserv.univ-lyon1.fr http://biomserv.univ-lyon1.fr/~pelosse/ Hello Griet, THE person to ask would be Patrick Meirmans, you'll find his contact info here: http://www.unil.ch/dee/page56674_en.html I think Spagedi may be able to deal with diploids and triploids... Good luck, Marc. ==========!!! PLEASE NOTE ROOM AND PHONE NUMBER CHANGE!!! ==========Marc Stift Division of Ecology & Evolutionary Biology Graham Kerr Building Molecular Ecology Group (Room 411) Graham Kerr Building University of Glasgow Glasgow G12 8QQ, SCOTLAND Office: +44 (0)141 330 6625 m.stift@bio.gla.ac.uk Dear colleague, I would be interested to get the answers to your request. Thanks very much in advance... Matthias Stöck I suggest you write to Brian Husband at University of Guelph. When you email him cc Paul Krohn his research associate. Marc Hi, It's not exactly what you want (it doesn't do assignment) but it may be helpful to you to know that the LAMARC program, which estimates population parameters from molecular data, can handle triploids and mixed diploids and triploids. The person who programmed this capability has been hoping to see some triploids ever since! Best wishes, Mary Kuhner mkkuhner@u.washington.edu http://evolution.gs.washington.edu/lamarc.html Dear Griet, Alas, I am afraid I do not have an answer for your, only a further question. I also work with a species complex (of leaf beetles) that includes triploids along with diploids. We are fortunate in that the focal taxa on which our research program primarily focuses have diploids and triploids that are distinguishable on the basis of a single mtDNA restriction site. However, this approach is not as reliable for some of the other species. Another approach we have taken is flow cytometry but our experience in collaborating with someone on this is that it is a relatively slow and expensive approach. Thus, my question to you is whether you use or know of other means of molecularly distinguishing diploid from triploid individuals. Thanks, and good luck receiving useful replies to your e-mail. Best, Dan Daniel J. Funk Associate Professor Dept. of Biological Sciences Vanderbilt University Nashville, TN 37235 615-322-2214 Hello Griet, If you have access to a Mac, you can use my own program GenoDive, that can do Amova's, Mantel tests, clustering, PCA and other stuff. It also has some special features for analysing data from asexuals, which may come in handy when you are working with Rana esculenta (why did they change the name to the horrible sounding "Pelophylax"?). GenoDive can be downloaded from: http://www.patrickmeirmans.com/software If you are working with a PC, Spagedi by Olivier Hardy and Xavier Vekemans is a very good program that is mostly used for spatial analyses, but can also do normal stuff like F_st and calculate the ploidy independent F_st analogue Rho_st. Of course Structure can also work with polyploid data, but it assumes random mating, which may not be the case in your study species. If you have any further questions, don't hesitate to ask. Cheers, Patrick p.s. the link to your website in the signature of your email does not work... After more communication I received my solution: Hello Griet, Structure can very well handle triploid allelic data like microsatellites. So it is not necessary to convert your data to AFLPs before the analysis, this will only lead to a loss of information, and therefore less reliable results. Structure can also handle mixtures of different ploidy levels. I did this myself for a mixture of dandelions of different ploidy levels. The trick is that you have to code the diploids as being triploid with one missing allele at every locus. If you have some loci with more than three alleles, you can do the same trick by filling all loci for all individuals up to the maximum number of alleles that you found at a single locus. So maybe you have to make them all pentaploid if you found a maximum of five alleles. I did some tests on this by coding the data in different ways and also using simulated data, and found that there is no bias arising from all the missing data, so the missing data is really discarded. I attached my own datafile in order to show you the principle. All individuals labeled 0 in the second column are diploids and the ones labeled 1 are triploids. Hope this helps, Patrick -- Drs. Griet Holsbeek Katholieke Universiteit Leuven Laboratory of Aquatic Ecology Ch. de Beriotstraat, 32 B-3000 Leuven Belgium Phone: +32 16 32 39 66 (secretariat) or +32 16 32 45 72 Fax: +32 16 32 45 75 E-mail: griet.holsbeek@bio.kuleuven.be website: http://bio.kuleuven.be/de/dea/ Griet Holsbeek