================================================= Release notes for OMA standalone 1.x/2.x series ================================================= :Author: OMA Developers :Contact: contact@omabrowser.org Summary of Changes from 2.5.0 to 2.6.0 ====================================== New features: ------------- - produce orthoxml files according to v0.5 of orthoxml (https://github.com/qfo/orthoxml/releases/tag/v0.5) - Map-SeqNum-ID.txt contains an additional column indicating which proteins were selected as main isoforms. Bug Fixes: ---------- - exclude all but main isoform variant from the HierarchicalGroups.orthoxml output (https://github.com/DessimozLab/OmaStandalone/issues/8) - fixes an issue with parsing splicing information for genomes with many splicing variants. - minor improvements in the documentation. Summary of Changes from 2.4.2 to 2.5.0 ====================================== New features: ------------- - introduces new variable AuxDataPath in parameters file that allows to specify the path where auxillary datafiles are stored. - provide docker container on dockerhub for OMA standalone. dessimozlab/oma_standalone Bug Fixes: ---------- - fixes problems running OMA standalone on debian buster clusters (gh-issue #3) - minor improvements in the documentation. Summary of Changes from 2.4.1 to 2.4.2 ====================================== This is a bugfix release only. Bug Fixes --------- - fixes problem in bottom-up HOG inference with long protein fasta headers (see https://www.biostars.org/p/477881) - minor improvements in the documentation Summary of Changes from 2.4.0 to 2.4.1 ====================================== This is a bugfix release only. Bug Fixes --------- - fixes issue that prevents installation via homebrew/linuxbrew - additional checks during installation - minor improvements in the documentation Summary of Changes from 2.3.1 to 2.4.0 ====================================== This release contains a few new features and bug fixes. New Features: ------------- - Bottom-up algorithm is now the default algorithm for infering HOGs. This variant is better and more scaleable as the original top-down variant. The top-down can still be used by setting the DoHierarchicalGroups parameter is set to 'top-down' in the parameters file. - Various performance improvements Bug Fixes --------- - fixes a problem in oma-cleanup that removes left ckpt files. - fixes various typos, improvements in documentation - bottom_up procedure for HOGs keep the file of the exact used species tree for all possible parameters for `SpeciesTree` parameter. Summary of Changes from 2.3.0 to 2.3.1 ====================================== This is a bugfix release only. Bug Fixes --------- - bottom-up HOG output file contains species name in species tag - fixes url to retrieve GeneOntology definition - improve performance in reading input fasta files - improve error handling - fixes documentation typos Summary of Changes from 2.2.0 to 2.3.0 ====================================== New Features: ------------- - Added auxilary scripts oma-compact, oma-status, oma-cleanup to reduce nr of produced files, check overall progress of allall computations and cleaning up of files after crashes. - Estimated species tree is additionally stored in phyloxml format Bug Fixes --------- - fixes a bug in HOG inference bottom-up algorithm that led to bigger HOGs than intended for deep HOGs. With this fix we have also changed the default parameter MinEdgeCompletenessFraction. - changed exit code from 3 to 99 in case the specified max runtime exceeded. - fixes documentation typos - fixes bug with '/' and set TMPDIR environment variable in bottom-up GETHOGs. Thanks to Ferdinand Marletaz for reporting. - explicitly list duplicated protein IDs per genome in warning Summary of Changes from 2.1.1 to 2.2.0 ====================================== This release contains a few new features and bug fixes. Analyses obtained from previous OMA standalone versions are still fully compatible, so it is safe to carry over the Cache/ (and of course also DB/) folders. The new release contains a few more parameters. Unspecified parameters since 2.0.0 will quietly be set to their default value, but not stored in the parameter file. New Features: ------------- - produce phyletic profile matrices for OMA Groups and HOGs - additional parameters and options to better parametrize the prediction of GO functions from exported genomes Bug Fixes --------- - fixes various performance problems with checkpointing and pairs iterator that could lead to inf loop in short queues with automatic job resubmission - improved logging output - Fixing a few typos. - Improved parameter checks Summary of Changes from 2.1.0 to 2.1.1 ====================================== This is a bugfix release only. Bug Fixes --------- - fixed installer script to be compatible with homebrew. Summary of Changes from 2.0.0 to 2.1.0 ====================================== This release contains a few new features and several bug fixes. Analyses obtained from previous OMA standalone versions are still fully compatible, so it is safe to carry over the Cache/ (and of course also DB/) folders. New Features: ------------- - Checkpointing AllAll chunks before termination when specified max time is used. - create a new instance of a default parameter file. Bug Fixes --------- - fixed resolving the oma launch script on mac osx if called through a symlink. - Fixing a few typos. Summary of Changes from 1.1.2 to 2.0.0 ====================================== This is a major new release. We added several algorithmic improvements to OMA standalone, i.e. two changes in the pairwise orthology inference step and a new algorithm to infer the HOGs (bottom-up). The predicted orthologs will therefor likely be slightly different than in previous versions. The cached All-against-all is still fully compatible with the new release and does not need to be recomputed. New Features: ------------- - Allow to specify a time limit for the AllAll computations with a special exit code. - Improve rescuing of inparalogs where one copy evolves significantly faster than the other copy - Add an additional check for additivity of the distances when checking for hidden paralogy - add bottom-up HOG algorithm as an experimental feature. Bug Fixes --------- - improved stability for big job-arrays on HPC systems Summary of Changes from 1.1.1 to 1.1.2 ====================================== Bug Fixes --------- - Avoid printing warning when downloading Gene Ontology hierarchy for the first time. - Fixing a bug occuring on ubuntu with downloading data files. - Properly terminate OMA standalone runs at end except if run in interactive mode. - Extended documentation with more details about the output files. Summary of Changes from 1.1.0 to 1.1.1 ====================================== This is a bugfix release only. Bug Fixes --------- - Downloading files uses either wget or curl. If none of the two available, OMA ends with a clearer error message. - MinLength parameter does not discard short sequences. - Improving support on HPC platforms such that all children jobs are terminated if the job gets terminated. Summary of Changes from 1.0.6 to 1.1.0 ====================================== This release contains a few new features and several bug fixes. Analyses obtained from previous OMA standalone versions are still fully compatible, so it is safe to carry over the Cache/ (and of course also DB/) folders. New Features ------------ - option to merge Quasi-cliques into bigger OMA Groups - gene ontology function annotation based on OMA group membership and functions from exported genomes. - new parameter consistency checks for fast abort and detection of cache inconsistencies. Bug Fixes --------- - properly escape IDs in orthoxml files - sequences longer than 100k residues/nucs can now be analysed - short sequences will not be discarded a priori - won't break with very long splicing maps (thanks to Javier Herrero) - fixing several spelling errors Summary of Changes from 1.0.5 to 1.0.6 ====================================== This is a bugfix release only. Bug Fixes --------- - ignore .DS_store files in Cache folder - improve a few messages to make problem better understandable. - fixes a critical bug introduced in 1.0.5 that ended up in deadlocks in some situations on HPC systems. - fixes a slurm specific problem with jobarrays - genomes exported from OMA browser produced slightly different ids in the Map-SeqNum-ID.txt file. Also, the for the orthoxml output files, specific types have been guessed by default. The new default is to use the same ids as in the fasta header for all genomes. Without adjusting the parameters.drw file, still the old behaviour is mimicked. Summary of Changes from 1.0.4 to 1.0.5 ====================================== This is a bugfix release only. Bug Fixes --------- - avoid potential dataraces in situations where oma is started after changes of cached databases by removing a possible signaling file. Summary of Changes from 1.0.3 to 1.0.4 ====================================== This is a bugfix release only. Bug Fixes --------- - fixes a bug with the splicing variant data format in the .splice files which was not fully compatible with the expected format. There is now also an example .splice file for the ToyExample database of human. Summary of Changes from 1.0.2 to 1.0.3 ====================================== This is a bugfix release only. Bug Fixes --------- - fixes a bug introduced in 1.0.2 that does not respect genome order when looking for exported genomes AllAll files - include these release_notes in install tarball Summary of Changes from 1.0.1 to 1.0.2 ====================================== This is a bugfix release only. Bug Fixes --------- - fixes excessive memory consumption during the allall phase. - correct timing estimates during allall computations - updated manual text and scientific references Summary of Changes from 1.0.0 to 1.0.1 ====================================== This is a bugfix release only. Bug Fixes --------- - fixed listing of unfinished AllAll jobs in cases when the list is long. - too many unknown AA in a sequence will properly skip the sequence with a warning message on stderr. Summary of Changes from 0.99v to 1.0.0 ====================================== With the release of OMA standalone 1.x series we switch from a beta to a stable software product. Additional features ------------------- - Support of alternative splicing variants as in public OMA database. The strategy is described in OMA NAR 2010. - Additional support of HPC schedulers: - SLURM - PBS - made output folder a parameter - command line flag to stop after all-against-all phase. This allows different resource allocations for subprocesses on HPC. - command line flag to drop into interactive session on error - support to infer orthology based on cDNA sequences Improvements ------------ - Checks to verify consitency between genomes and cached results - store all-against-all files in sub-folders to avoid too many files per directory. - possibility to avoid storing output files that are not needed and can take a lot of time to compute. - extended documentation - moved to some more sane release numbers :-) Bug Fixes --------- - stable genome order accross filesystems - skip duplicated entries in AllAll files - proper conversion of DNA nucleotides