We will be a mothur workshop in December. Learn more.
We are happy to announce the release of mothur v.1.14 in time for Halloween - trick or treat!
This is update has a number of new commands and options that are just plain awesome. I'll call your attention to three that we are particularly excited about. The first is a command for the complete pipeline that the Schloss lab uses to process their own data (pipeline.pds). This command is written so that you can modify the pipeline to your lab's unique needs. The second is a wrapper for the metastats code that was generously provided by Mihai Pop's research group. The third will normalize your shared files so that each group has the same number of sequences. In the next release we hope to do a similar type of processing by randomly sub-sampling sequences from each group; however, we think the normalize.shared approach is better. Among the new options that are available you'll notice that we added parallelization to three more commands. The feature that many people will appreciate most are the addition of the optimize and criterion options to screen.seqs. These options should make running the command much easier (and limit the number related emails that pile up in our inbox). A number of the new commands, options, and bug fixes were suggested by users of the online mothur forum as well as those contacting us via email@example.com. We really appreciate this input and thank you for helping us to make mothur a great program.
Going forward we are excited about the features we have planned for the next release, so stay tuned. You have probably noticed that the number of new features and options continues to grow. This is awesome, but at the same time it seems as though there are two other commands we'd like to develop and test for every one we put into mothur. Because of this we are looking for someone to fill a postdoc-type position in the Schloss lab at The University of Michigan. If you'd like to learn more, please contact Pat at pschloss @ umich.edu. The basic qualifications are 1) good social skills, 2) knowledge of some programming language (preferably C++), 3) knowledge of microbial ecology, 4) good social skills.
As always, happy writing and let us know if we can be of assistance. Most importantly, keep the six packs coming...
- pipeline.pds - an implementation of Pat's pipeline in a single command, plus the ability to customize
- metastats - a wrapper for the metastats program from Mihai Pop's research group
- normalize.shared - normalize the number of sequences per group to a specified level
- split.groups - generates a fasta file for each group in a groupfile
- cluster.fragments - a modified version of pre.cluster / unique.seqs to cluster aligned sequences that overlap, but aren't exact
- get.lineage / remove.lineage - similar to remove.seqs/get.seqs except that instead of taking in an accnos file, one would give a taxonomy file and the name of a taxonomic group
- fastq.info - generate fasta and qual files from a fastq file
- deunique.seqs - re-replicate duplicate sequences
- pairwise.seqs - calculate pairwise distances between sequences without using a multiple-sequence alignment
- screen.seqs - added optimize and criteria options
- rarefaction.single - added processors option
- phylo.diversity - added processors option
- summary.shared - added processors and distance options
- summary.single - groupmode=T is the new default
- venn - added nseqs and permute options
- chop.seqs - added short option
- mothur will now look in the output directory automatically if it cannot find your file in the default locations, so if you are redirecting your output, you can still use simple names as inputs.
- when you go from heatmap.bin() to heatmap.sim() we get a segfault
- bug in 1.13.0 with shared files smart distancing
- if using make.group with more then 3 fasta files, resulting file will be called merge.groups
- fixed the unifrac commands so that their output order matches summary commands
- fixed bug introduced in version 1.13.0 that produced significance scores of 1 for unifrac.unweighted
- fixed bug where chimera commands didn't output to the correct folder if you used set.dir or a complete file name.
- classify.otu should put in "no_consensus" if below threshold
- fixed bug with unifrac.weighted, unifrac.unweighted and phylo.diversity that mistakenly added the root to the branch length
- output of pcoa command changed for *.pca to *.pcoa and *.pca.loadings to *pcoa.loadings