QIIME 1.9.1

This is a bug fix release that addresses three critical bugs, as well as several minor bugs and usability enhancements. The full details of the changes in this release, including descriptions of the three critical bugs, are in the ChangeLog. We apologize for any inconvenience that this may have caused.

We recommend that all users review the ChangeLog to determine if you were affected by these bugs. We recommend that all users upgrade from QIIME 1.9.0 to QIIME 1.9.1. See the instructions for upgrading to the latest version of QIIME.

QIIME 1.9.0

We're excited to announce QIIME 1.9.0. As is typical for our releases, the QIIME 1.9.0 Virtual Box and EC2 image will be ready in about a week.

Installing QIIME 1.9.0

The easiest way to install QIIME is by using pip. You can find the instructions for doing this in our (completely updated!) install document. See our updated upgrade documentation if you'd like to upgrade a pre-existing QIIME installation.

About QIIME 1.9.0

QIIME 1.9.0 contains a huge amount of new features. Some of the highlights include:

• Beta support has been added for open source alternatives for all OTU picking algorithms (de novo, closed-reference, and open-reference, including subsampled open-reference). These make use of SortMeRNA for closed-reference steps, and swarm and SumaClust for de novo steps. Open reference OTU picking with these tools can be accessed with pick_open_reference_otus.py -m sortmerna_sumaclust.
• Added three new workflow scripts for facilitating initial QIIME processing of already-demultiplexed fastq files, as these are commonly being provided by sequencing centers. These are: multiple_split_libraries_fastq.py, multiple_join_paired_ends.py, and multiple_extract_barcodes.py. We've re-written our Illumina Processing Documentation to describe these, and some other scripts, that will help you process raw Illumina data.
• Added new observation_metadata_correlation.py script. This script allows the calculation of correlations between feature abundances and continuous-valued metadata. This script replaces the continuous-valued correlation functionality that was in otu_category_significance.py in QIIME 1.7.0 and earlier.
• Added new compute_taxonomy_ratios.py script, which implements the microbial dysbiosis index (MD-index) from Gevers et al 2014.
• Added collapse_samples.py, which can be used for collapsing groups of samples in BIOM tables and mapping files based on their metadata (see #1678). This can be used, for example, to collapse samples belonging to a replicate group. This also has replaced summarize_otu_by_cat.py (see discussion on #1798).
• Added differential_abundance.py to supplement group_significance.py which supports metagenomeSeq's fitZIG algorithm and DESeq2's negative binomial algorithm. Similarly, we added normalize_table.py to support normalization algorithms in addition to rarefaction. Supported methods are metagenomeSeq's CSS and DESeq's variance stabilizing transformation.
• Added script compare_trajectories.py, which provides access to analysis of volatility using different algorithms.
• Added script start_parallel_jobs_slurm.py, which allows for parallel job submission using slurm.
• Updated a lot of our web documentation, including the 454 Overview Tutorial, the Illumina Overview Tutorial, the EC2 tutorial, and the QIIME script index.
• Greengenes 13_8 is now installed as part of the QIIME base install, and the 97% reference OTUs are used as the default reference database for the OTU pickers and taxonomy assigners. This is convenient for users working with 16S data, and can easily be overridden for users working with other marker genes. This means that all of the QIIME workflows can be run immediately after pip installing QIIME - there is no need to download other files or create a qiime_config.

This is just a sneak-peek at some of the new features that are packed into QIIME 1.9.0. See the ChangeLog for a lot more detail.

Finally, thanks to all of the QIIME developers for the huge amount of effort that went into this new release, and to our users for using QIIME and for testing the 1.9.0 release candidates.

QIIME 1.9.0 release candidate 2

We're excited to announce the second QIIME 1.9.0 release candidate: QIIME 1.9.0-rc2.

About the release candidate

This is the first time we're preparing a candidate release in advance of an official QIIME release, but it's likely a pattern that we'll follow in the future. We're hoping that users will begin experimenting with QIIME 1.9.0-rc2, in parallel with the QIIME developers doing extensive stress testing of the new functionality.

The actual QIIME 1.9.0 release will follow in about two weeks (target date: 21 January 2015). The only planned changes between now and the official release are improvements to the documentation (and the tutorials in particular, which need some updating to cover 1.9.0); additions to test code as necessary (e.g., if we identify a problem with the release candidate, we'll add tests that detect that problem); and bug fixes as necessary. No new functionality will be added between QIIME 1.9.0-rc2 and 1.9.0.

At this time, we're not updating the QIIME website, and we're not providing new virtual machine images for AWS and Virtual Box. The website will be updated when QIIME 1.9.0 is released, and the virtual machine images will follow a couple of days later (as they typically do).

If you identify problems with QIIME 1.9.0-rc2

Our goal for the release candidate is to find and fix bugs that otherwise would have been found post-release. If you identify issues with QIIME 1.9.0-rc2, please post them to the QIIME Issue Tracker (you'll need a free GitHub account to do this) with as much detail as possible.

Installing the release candidate

The easiest way to install the release candidate is by running:

pip install numpy
pip install https://github.com/biocore/qiime/archive/1.9.0-rc2.tar.gz

This will give you a fully working QIIME base install. You can then test this install by running:

print_qiime_config.py -t

You can read more about installing QIIME in our (completely updated!) install document. (Note: you'll be viewing the install document on the GitHub website when you view it from that link. Some parts won't be rendered correctly, but that will be fixed when this is published to the QIIME website on release of QIIME 1.9.0.)

About QIIME 1.9.0

QIIME 1.9.0, and therefore this release candidate, contain a huge amount of new features. Some of the highlights include:

• Beta support has been added for open source alternatives for all OTU picking algorithms (de novo, closed-reference, and open-reference, including subsampled open-reference). These make use of SortMeRNA for closed-reference steps, and swarm and SumaClust for de novo steps. Open reference OTU picking with these tools can be accessed with pick_open_reference_otus.py -m sortmerna_sumaclust.
• Added three new workflow scripts for facilitating initial QIIME processing of already-demultiplexed fastq files, as these are commonly being provided by sequencing centers. These are: multiple_split_libraries_fastq.py, multiple_join_paired_ends.py, and multiple_extract_barcodes.py.
• Added new observation_metadata_correlation.py script. This script allows the calculation of correlations between feature abundances and continuous-valued metadata. This script replaces the continuous-valued correlation functionality that was in otu_category_significance.py in QIIME 1.7.0 and earlier.
• Added new compute_taxonomy_ratios.py script, which implements the microbial dysbiosis index (MD-index) from Gevers et al 2014.
• Added collapse_samples.py, which can be used for collapsing groups of samples in BIOM tables and mapping files based on their metadata (see #1678). This can be used, for example, to collapse samples belonging to a replicate group. This also has replaced summarize_otu_by_cat.py (see discussion on #1798).
• Added differential_abundance.py to supplement group_significance.py which supports metagenomeSeq's fitZIG algorithm and DESeq2's negative binomial algorithm. Similarly, we added normalize_table.py to support normalization algorithms in addition to rarefaction. Supported methods are metagenomeSeq's CSS and DESeq's variance stabilizing transformation.
• Added script compare_trajectories.py, which provides access to analysis of volatility using different algorithms.
• Added script start_parallel_jobs_slurm.py, which allows for parallel job submission using slurm.

This is just a sneak-peek at some of the new features that are packed into QIIME 1.9.0-rc2. See the ChangeLog, which we'll be finalizing before QIIME 1.9.0, for a lot more detail.

QIIME 1.9.0 release candidate 1

Please use QIIME 1.9.0-rc2 for testing

QIIME 1.9.0-rc2 has now been released. All testing should be done using that release, rather than QIIME 1.9.0-rc1.

QIIME 1.9.0-rc1

We're excited to announce the first QIIME 1.9.0 release candidate: QIIME 1.9.0-rc1.

About the release candidate

This is the first time we're preparing a candidate release in advance of an official QIIME release, but it's likely a pattern that we'll follow in the future. We're hoping that users will begin experimenting with QIIME 1.9.0-rc1, in parallel with the QIIME developers doing extensive stress testing of the new functionality.

The actual QIIME 1.9.0 release will follow in about one month, though we don't have the specific date set yet. The only planned changes between now and the official release are improvements to the documentation (and the tutorials in particular, which need some updating to cover 1.9.0); additions to test code as necessary (e.g., if we identify a problem with the release candidate, we'll add tests that detect that problem); and bug fixes as necessary. No new functionality will be added between QIIME 1.9.0-rc1 and 1.9.0.

At this time, we're not updating the QIIME website, and we're not providing new virtual machine images for AWS and Virtual Box. The website will be updated when QIIME 1.9.0 is released, and the virtual machine images will follow a couple of days later (as they typically do).

If you identify problems with QIIME 1.9.0-rc1

Our goal for the release candidate is to find and fix bugs that otherwise would have been found post-release. If you identify issues with QIIME 1.9.0-rc1, please post them to the QIIME Issue Tracker (you'll need a free GitHub account to do this) with as much detail as possible.

Installing the release candidate

The easiest way to install the release candidate is by running:

pip install numpy
pip install https://github.com/biocore/qiime/archive/1.9.0-rc1.tar.gz

This will give you a fully working QIIME base install. You can then test this install by running:

print_qiime_config.py -t

You can read more about installing QIIME in our (completely updated!) install document. (Note: you'll be viewing the install document on the GitHub website when you view it from that link. Some parts won't be rendered correctly, but that will be fixed when this is published to the QIIME website on release of QIIME 1.9.0.)

About QIIME 1.9.0

QIIME 1.9.0, and therefore this release candidate, contain a huge amount of new features. Some of the highlights include:

• Beta support has been added for open source alternatives for all OTU picking algorithms (de novo, closed-reference, and open-reference, including subsampled open-reference). These make use of SortMeRNA for closed-reference steps, and swarm and SumaClust for de novo steps. Open reference OTU picking with these tools can be accessed with pick_open_reference_otus.py -m sortmerna_sumaclust.
• Added three new workflow scripts for facilitating initial QIIME processing of already-demultiplexed fastq files, as these are commonly being provided by sequencing centers. These are: multiple_split_libraries_fastq.py, multiple_join_paired_ends.py, and multiple_extract_barcodes.py.
• Added new observation_metadata_correlation.py script. This script allows the calculation of correlations between feature abundances and continuous-valued metadata. This script replaces the continuous-valued correlation functionality that was in otu_category_significance.py in QIIME 1.7.0 and earlier.
• Added new compute_taxonomy_ratios.py script, which implements the microbial dysbiosis index (MD-index) from Gevers et al 2014.
• Added collapse_samples.py, which can be used for collapsing groups of samples in BIOM tables and mapping files based on their metadata (see #1678). This can be used, for example, to collapse samples belonging to a replicate group. This also has replaced summarize_otu_by_cat.py (see discussion on #1798).
• Added differential_abundance.py to supplement group_significance.py which supports metagenomeSeq's fitZIG algorithm and DESeq2's negative binomial algorithm. Similarly, we added normalize_table.py to support normalization algorithms in addition to rarefaction. Supported methods are metagenomeSeq's CSS and DESeq's variance stabilizing transformation.
• Added script compare_trajectories.py, which provides access to analysis of volatility using different algorithms.
• Added script start_parallel_jobs_slurm.py, which allows for parallel job submission using slurm.

This is just a sneak-peek at some of the new features that are packed into QIIME 1.9.0-rc1. See the ChangeLog, which we'll be finalizing before QIIME 1.9.0, for a lot more detail.