Hello, I am relatively new to vegan and need some help with metaMDS. My R is Version 3.0.2. I am analysing cuticular profiles of flys from different locations an different gender. My data is in a 48 x 70 matrix. It is about percentages and so, there are a lot of 0s included. If I run metaMDS, it just gives me: Warning message: In metaMDS(data1, distance = "bray", k = 2, trymax = 100, autotransform = F) : Stress is (nearly) zero - you may have insufficient data no matter, what I change. The error is always around 9.241406e-05, which seems to be far to low for my data. By inspecting the data sheet, I clearly see that there must be two clusters ? but there are still some differences in these (meaning: Not all female flies show the same pattern than the others and there are slight differences in the locations ?) - so I expect to see two clusters for the sexes and in these, an accumulation of the different locations. But metaMDS just gives me 2 points (with ordiplot), where the 24 males and the 24 females are completely stacked. That's just some kind of worthless for me. So is there any chance to get some nice clusters? I tried to limit the number of iterationswith maxit, but had to go down to 10 to reach a suitable result. And I don't think, I may limit the number of iterations that much. What could possibly be the problem? It would be great if somebody could help me. Thaks! Fabian -- View this message in context: http://r.789695.n4.nabble.com/Problem-with-metaMDS-in-vegan-tp4685022.html Sent from the R help mailing list archive at Nabble.com.
Problem with metaMDS in vegan
6 messages · David L Carlson, Jari Oksanen, Fabian B
Without more information, it is hard to say. You did not tell us much about the data beyond the dimensions, but it looks like you have several different kinds of measurements including location and gender (probably categorical) and cuticular profiles (presumably numeric). Without the commands you used, we can only guess what you are doing, but the snippet you provided suggests that you dumped different data types into metaMDS and used the Bray distance coefficient. The Bray coefficient is designed for analyzing community structure by comparing the counts of different species between samples. It does not appear that this is what you are trying to do. ------------------------------------- David L Carlson Department of Anthropology Texas A&M University College Station, TX 77840-4352 -----Original Message----- From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of Fabian B Sent: Monday, February 10, 2014 5:27 AM To: r-help at r-project.org Subject: [R] Problem with metaMDS in vegan Hello, I am relatively new to vegan and need some help with metaMDS. My R is Version 3.0.2. I am analysing cuticular profiles of flys from different locations an different gender. My data is in a 48 x 70 matrix. It is about percentages and so, there are a lot of 0s included. If I run metaMDS, it just gives me: Warning message: In metaMDS(data1, distance = "bray", k = 2, trymax = 100, autotransform = F) : Stress is (nearly) zero - you may have insufficient data no matter, what I change. The error is always around 9.241406e-05, which seems to be far to low for my data. By inspecting the data sheet, I clearly see that there must be two clusters ? but there are still some differences in these (meaning: Not all female flies show the same pattern than the others and there are slight differences in the locations ?) - so I expect to see two clusters for the sexes and in these, an accumulation of the different locations. But metaMDS just gives me 2 points (with ordiplot), where the 24 males and the 24 females are completely stacked. That's just some kind of worthless for me. So is there any chance to get some nice clusters? I tried to limit the number of iterationswith maxit, but had to go down to 10 to reach a suitable result. And I don't think, I may limit the number of iterations that much. What could possibly be the problem? It would be great if somebody could help me. Thaks! Fabian -- View this message in context: http://r.789695.n4.nabble.com/Problem-with-metaMDS-in-vegan-tp4685022.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Thanks for the fast response! My data is actually just the percentages of the substances found in the cuticular layer of the flies. The quantitative and qualitative composition of this layer differs in gender, locality and so on. My 70 ?species? are the chemical compounds we identified and my 48 ?sites? are the individuals (location and gender are encrypted in the individuals' recognition code). Therefore, I have the percentages of one chemical compound over all tested individuals in one row (as numeric data). You could say that these are the abundances of the species (the compounds) at the different sites (individuals). So my data is quite similar to community structure. I have done this before and it worked just fine. I used: metaMDS(data, distance=?bray?, k=2, trymax=10000, autotransform=F) As I mentioned before, this time the stress is very low (around 9.5 * 10^-5) and R gives me that warning message. Though I have always had data in the same structure, I've never been in this situation. The NMDS just stops after the first run, because of the low stress. But there are obviously some differences in my data ? I can see them with my eye. If I discard the 'distance=?bray?', nothing changes. May it be possible that these are simply just to little to be recognised correctly? How could I change that? If I add 'maxit=10', I get quite what I assumed. Two completely separated clusters with cloudlike appearance (stress around 0.02). The clusters are the two sexes and in each cluster, the localities group together. But I'm not quite sure, if I may set a maxit-level that low ? or if I just create some false relations with that. I also recognised , that the stress gets lower, the higher I set 'maxit' and the lower I set 'trymax'. What could be going wrong? I really have no clue ? Thanks for your help! Regards, Fabian -- View this message in context: http://r.789695.n4.nabble.com/Problem-with-metaMDS-in-vegan-tp4685022p4685043.html Sent from the R help mailing list archive at Nabble.com.
A reproducible example would help. Look at the example in metaMDS using the dune data set. It consists of 20 observations (sites) with 30 species observed (columns). You indicate that your compounds (species) are the rows and your sites are the columns. So perhaps you are analyzing the distances between compounds not between sites? Have you tried analyzing t(data)? David C -----Original Message----- From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of Fabian B Sent: Monday, February 10, 2014 9:58 AM To: r-help at r-project.org Subject: Re: [R] Problem with metaMDS in vegan Thanks for the fast response! My data is actually just the percentages of the substances found in the cuticular layer of the flies. The quantitative and qualitative composition of this layer differs in gender, locality and so on. My 70 ?species? are the chemical compounds we identified and my 48 ?sites? are the individuals (location and gender are encrypted in the individuals' recognition code). Therefore, I have the percentages of one chemical compound over all tested individuals in one row (as numeric data). You could say that these are the abundances of the species (the compounds) at the different sites (individuals). So my data is quite similar to community structure. I have done this before and it worked just fine. I used: metaMDS(data, distance=?bray?, k=2, trymax=10000, autotransform=F) As I mentioned before, this time the stress is very low (around 9.5 * 10^-5) and R gives me that warning message. Though I have always had data in the same structure, I've never been in this situation. The NMDS just stops after the first run, because of the low stress. But there are obviously some differences in my data ? I can see them with my eye. If I discard the 'distance=?bray?', nothing changes. May it be possible that these are simply just to little to be recognised correctly? How could I change that? If I add 'maxit=10', I get quite what I assumed. Two completely separated clusters with cloudlike appearance (stress around 0.02). The clusters are the two sexes and in each cluster, the localities group together. But I'm not quite sure, if I may set a maxit-level that low ? or if I just create some false relations with that. I also recognised , that the stress gets lower, the higher I set 'maxit' and the lower I set 'trymax'. What could be going wrong? I really have no clue ? Thanks for your help! Regards, Fabian -- View this message in context: http://r.789695.n4.nabble.com/Problem-with-metaMDS-in-vegan-tp4685022p4685043.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Fabian B <fabian.boetzl <at> freenet.de> writes:
Thanks for the fast response! My data is actually just the percentages of the substances found in the cuticular layer of the flies. The quantitative and qualitative composition of this layer differs in gender, locality and so on. My 70 ?species? are the chemical compounds we identified and my 48 ?sites? are the individuals (location and gender are encrypted in the individuals' recognition code). Therefore, I have the percentages of one chemical compound over all tested individuals in one row (as numeric data). You could say that these are the abundances of the species (the compounds) at the different sites (individuals). So my data is quite similar to community structure. I have done this before and it worked just fine. I used: metaMDS(data, distance=?bray?, k=2, trymax=10000, autotransform=F) As I mentioned before, this time the stress is very low (around 9.5 * 10^-5) and R gives me that warning message. Though I have always had data in the same structure, I've never been in this situation. The NMDS just stops after the first run, because of the low stress. But there are obviously some differences in my data ? I can see them with my eye. If I discard the 'distance=?bray?', nothing changes. May it be possible that these are simply just to little to be recognised correctly? How could I change that? If I add 'maxit=10', I get quite what I assumed. Two completely separated clusters with cloudlike appearance (stress around 0.02). The clusters are the two sexes and in each cluster, the localities group together. But I'm not quite sure, if I may set a maxit-level that low ? or if I just create some false relations with that. I also recognised , that the stress gets lower, the higher I set 'maxit' and the lower I set 'trymax'. What could be going wrong? I really have no clue ?
I think nothing is going wrong: you just have that kind of data, and there is nothing wrong there. There is nothing to reproduce here, but it *seems* that you may have a disjoint data set with two clusters sharing nothing between them. The metaMDS is obsessed with that difference and does its best to separate these clusters. If you set 'maxit=10', it does not have time to finish the job, and you get two clouds. Just let it finish, and the clouds are squeezed to singularities. If you want to see the structure within these singularities, you could analyses these groups of points separately. Then the major difference between clouds won't obscure you. I am not quite convinced that Bray-Curtis (distance='bray') is the best choice of dissimilarity measure for these kind of data. If you have scaled values (percentages), choices like 'distance = "manhattan"' or 'distance = "euclidean"' or 'distance = "gower"' could work better. As we have no reproducible example, this is all guessing. Cheers, Jari Oksanen
Thank you very much! This is exactly the problem, I am in. I ran the NMDS for only one gender and it shows me the results, I was expecting. I was told to use bray-curtis, because there are a lot of "0"s in my data set and in e.g. manhattan these would have to much influence on the result. But in nature, chemicals that are only weakly represented may not have very much influence on the other individuals. For the moment, I won't question that decision, because it's quite a convention in chemical ecology, as I heard. Regards, Fabian -- View this message in context: http://r.789695.n4.nabble.com/Problem-with-metaMDS-in-vegan-tp4685022p4685134.html Sent from the R help mailing list archive at Nabble.com.