I am new to Mothur so I have been tutoring myself using the very explanatory manual for Mothur. However, I am having a bit of trouble. I am trying to determine the OTUs for my set of around 500 sequences which encompasses sequences from various sampling sites. I have a FASTA file of the alignments I have made and I built a distance matrix using the dist.seqs command. When I read the distance matrix on to the memory and cluster it though I get no OTUs. I am wondering if it is something I am doing wrong. I would be very grateful for any help or insights in to this.
Nikhil Ram Mohan
Department of Molecular and Cell Biology
University of Connecticut
Can you give us a blow by blow list of the commands you are using so we can help you out? Have you looked at the Costello analysis example to see how to proceed?
Thanks a lot for offering to help. This is how I went about it.
I have an alignment of around 500 sequences that I created using MacClade. I created a unique file on these using the unique.seqs command. Next I filtered the sequences before I could create a distance matrix using the dist.seqs command. I read the distance on to the memory using the read.dist command and then clustered it. This is what I did but got no OTUs.
I guess I was looking for a bit more detail. Are these 16S sequences? If so, why macclade? How are you filtering the sequences? When you look at the output of filtering are there any sequences. When you look at the distance matrix, are there any distances? If so, what is the smallest distance you see?
I am working with bacteriorhodopsin sequences from Haloarchaea in different Saltern lakes around the world. MacClade was my option to do the alignment because I wanted to do an amino acid level alignment. However, the FASTA file I am working with is an exported version of the nucleotide alignment. I checked the output files for the filter.seqs and the dist.seqs command and there are results in both of them. The lowest value in the distance matrix was 0 and there were a few that were around 0.003. The highest value I found was around 0.48. I was wondering if the filtering of sequences did have a role in me not getting any OTUs so I tried running the distance matrix without creating a unique file and filtering the sequences but without a result still.
I hope this is what you were looking for and I am truly very grateful for the help you are providing me.
Ok, that makes sense. Do your sequences all overlap the same region completely (and nothing extra?), i.e. did you use trump=. in filter.seqs? Feel free to email your alignment file to us at mother.bugs@gmail
I got it to work! I tried the different output formats for the distance matrix and for some reason only the ‘square’ distance matrix seemed to work for me. Thanks for all your help.