mothur v.1.17.0

February may be the shortest month of the year, but that doesn’t keep us from releasing two updates in the month! We are happy to announce the release of mothur v.1.17.0. We are excited about the mothur workshop in Detroit next week and are already planning for the next workshop towards the end of April, if you are interested in learning more, please let me know.

This release has several new commands for performing hypothesis testing to compare groups of samples including commands for performing the analysis of molecular variance, test of homogeneity of variance, analysis of similarity, and Mantel’s test. We previously discussed using AMOVA and HOMOVA with DNA sequences and are finally bringing that functionality to mothur for comparing groups of samples.

Another important facet of this release relates to fixing a number of critical bugs in how we were calculating the unifrac distances. If you have previously used mothur’s calculation of these values, please re-calculate them with the latest version. We are unsure how significant these bugs were to to people’s analyses. Needless to say, we apologize for any problems this may cause people and are very grateful to those that have brought these problems to our attention. If you know of anyone using mothur that isn’t receiving these emails, please spread the word.

Thanks to all of you that have completed our survey. If you haven’t completed it yet, there is still time...

New commands

  • amova - analysis of molecular variance
  • homova - test the homogeneity of variance
  • anosim - analysis of similarity
  • mantel - Mantel’s test of correlation between matrices
  • make.fastq - make your own fastq files

Feature updates

Bug fixes

  • Catchall command not finding files - https://forum.mothur.org/viewtopic.php?f=3&t=1114
  • sub.sample command that caused mothur to stall if size was close to total number of sequences in file.
  • unifrac commands - significance scores differing between multiple processors or one processors if the number of comparisons was less than the number of processors used.
  • unifrac commands that occurred if trees were unrooted.

Other stuff

  • added [error] flag when a command aborts