-
Fix #174.
-
Fix #173.
- Pygtftk now depends on Python 3.9.
- Force pillow version betwwen 7.0.0 and 8.3.2
-
Fix "Index.ravel() returning a np.ndarray is deprecated" in pandas.
-
Fix pillow requirements (8.4.0).
-
Add requirement for libjpeg-devel installation in build_wheels.sh (pillow requirement).
-
Force --no-cache when building docker image for manylinux package.
-
OLOGRAM README has been updated (add bedtools, graphviz and unzip dependency).
-
Replace operatorPrecedence with infixNotation when using pyparsing (see #166).
- Fix a bug in histogram drawing (OLOGRAM).
-
OLOGRAM doc has been updated.
-
Added test for Beta distribution fitting.
- OLOGRAM now gives the p-value for S as fitted through a Beta distribution. However, Beta distributions are harder to fit, since unavoidable small errors in the estimation of the skewness and kurtosis will propagate. In practice it is preferable to approximate this with a Negative Binomial. This will be explained in the manuscript.
- Fix an issue with manylinux (creation of pip releases). Another trial (see 1.2.9).
- Fix an issue with manylinux (creation of pip releases).
-
Fix gtf.tail() (#158).
-
Fix merge_attr (#147).
-
Fixed merged statistics calculation in ologram_merge_runs.
-
Fixed memory leaks in OLOGRAM's multiprocessing. RAM usage should decrease by 25-70% in certain cases.
-
Fixed a bug where OLOGRAM would needlessly store empty lists if certain combinations were not encountered.
-
Fixed several bugs in OLOGRAM that could result in dead process hanging indefinitely, including a histgram drawing bug.
-
OLOGRAM now merges consecutive intersections, to prevent a bug that resulted in double countings.
-
Fixed a bug in OLOGRAM where files with less than 2 regions were needlessly discarded.
-
OLOGRAM will now also produce separate histograms for each combination length.
-
The --chrom-file argument now accepts "mm8_ens", "mm9_ens", "mm10_ens", "hg19_ens", "hg38_ens", "rn3_ens" and "rn4_ens" which correspond to internal chromosome info files in ensembl format (i.e. no 'chr' prefix) (#146).
-
OLOGRAM has two new arguments (--hide-undef and --pval-threshold) to hide undefine or high p-values.
-
ologram_modl_treeify has a new argument (--min-inheritance) to add combinations only if they represent a given proportion of an existing parent.
-
In OLOGRAM-MODL, whether to use exact combinations or not (should A+B+C count as an observation of A+B ? Yes if not exact) is now controlled by an argument (--exact).
-
Added an argument to OLOGRAM to hide bars with undefined p-values (#149).
-
Minor optimizations to overlap computing.
-
pygtftk now requires Python >=3.8, <3.9.
-
Major optimizations to OLOGRAM's handling of RAM using newly-created structures.
-
Reduced OLOGRAM time cost by removing operations (computation of relative combination exactitude) that were not needed.
-
OLOGRAM's statistics computation with multiprocessing now uses a batch system instead of individual jobs.
-
OLOGRAM now has an argument to determine if certain sets within --more-beds should be kept intact during the shuffling.
-
The get_attr_value_list command now accepts -k '*'.
- Fixed LARS coder bug.
- Improved OLOGRAM documentation.
- Increased default alpha for MODL.
- Fix bioconda building for OSX.
- None.
- Fix requirements.txt
- None.
- Fixed Python version in Setup.py.
- None.
- Fixed duplicates in recursion.
- None.
-
Updated requirements.txt according to python 3.8.
-
MODL optional word normalization and smothering.
- Updated documentation
- Fix multiprocessing issue with py3.8.
- None.
- Updated requirements.txt according to python 3.8.
- None.
This version introduces OLOGRAM-MODL, a new paradigm for OLOGRAM to find intersections between multiple sets of genomic regions at once and then compute their enrichment with OLOGRAM. An optional algorithm (MODL) to find interesting combinations with sparse dictionary learning and greedy submodular optimisation has also been added. Furthermore, it also contains major speedups to OLOGRAM itself.
-
OLOGRAM - Fixed bug in multiple overlaps when using sets with no peak on a given chromosome.
-
Updated package requirements
-
OLOGRAM - The display graph has now the X labels at 90 degrees.
-
OLOGRAM - more-bed-labels should take and clean BED file names as default
-
OLOGRAM - Various graphical fixes
-
fix #124
-
fix BED to BED convertion in arg_formatted.FormattedFile(). BED6+ files were considered as BED6- files.
-
fix #136 although --show-group-number is no more supported with gtftk profile when plotnine > 0.6.0 is used.
-
Fix #128 (merge_attr error when using same key as source and destination).
-
Moved OLOGRAM-related commands to their own section in the documentation.
-
The MODL algorithm for combination mining can be accessed independantly.
-
Major speedups achieved in OLOGRAM by better typecasting in the Cython code.
-
Major speedup in OLOGRAM due to rewriting the pandas melt() function in C/Cython.
-
Added multithreading batch-by-batch for OLOGRAM
-
Renamed merge_ologram_stats to ologram_merge_stats.
-
Improved ologram_merge_stats visuals.
-
Added new simple_07 and ologram_2 example datasets to study multiple overlaps.
-
Added scikit-learn as a dependency.
-
Moved OLOGRAm functions to calculate enrichment to their own module.
-
This version implements OLOGRAM-MODL to study the enrichment of intersections between multiple sets of genomic regions. Please see the documentation and code comments for more details.
-
The API contains a Modl class which is a dictionary-learning based itemset mining algorithm, used in OLOGRAM-MODL
-
Introduced a treeify_ologram_modl plugin to visualize n-wise enrichment results as a treee
-
Introduced a ologram_merge_runs command to merge several runs to save RAM, treating each as a superbatch.
- None
- No more compatible with Python 3.5 (as BioPython).
- None.
- None.
- fix #123, #122, #121, #120
- None.
- None.
- None.
- None
- None.
- None.
- The --more-bed-labels is now facultative in OLOGRAM.
-
Fix #116 (pandas version issue)
-
Fix an issue related to pybedtool/bedtool version (naming of sequences that differs due to name/name+/nameOnly arguments).
-
Fix -n with integer values in get_5p_3p_coords.
- None.
- None.
-
md5sum-lite call have been replaced by "md5 -r" under darwin platforms.
-
The tss_numbering command now allows to add the number of different TSSs to the gene feature.
- None.
- None.
- None.
-
Support for Python 3.7.
-
The tss_numbering command now allows to add the number of different TSSs to the gene feature.
- None.
-
None.
-
bigwig_to_bed is know part of miscellaneous commands.
- The select_by_key command now accept seq_name and seqname in addition to seqid and chrom as a key.
-
This version contains the tss_numbering plugin. Annotate transcripts by computing their TSS position relative to the most five prime TSS of the corresponding gene.
-
The get_attr_value_list command now accepts a list of keys as input.
-
The get_attr_value_list command has an additional argument (--print-key-name).
-
Added a new miscellaneous command, get_ceas_record that extract records from CEAS (Cis-regulatory Element Annotation System).
-
Added a new miscellaneous command, great_reg_domains. This tool represents an attempt to process genomic annotations in GTF format in order to produced a set of 'labeled' regions with the same rules as those described in GREAT (Genomic Regions Enrichment of Annotations Tool) documentation. We can not warrant that the procedure is exactly the same. See the CLI for more details.
-
Added -y/--display-fit-quality to ologram
This version introduces ologram_merge_stats command that can be used to produce a heatmap from multiple OLOGRAM results.
- None.
- None.
- The select_by_key command now accept seq_name and seqname in addition to seqid and chrom as a key.
- This version implements ologram_merge_stats command that can be used to produce a heatmap from multiple OLOGRAM results.
This version contains some minor code refactoring. See 1.0.6 for recent major changes.
- None.
- None.
- Essentially some refactoring.
- None.
-
ologram warnings are now clearer.
-
Fixed reproducibility issue when using --more_keys in ologram.
-
Fix #78 (Relaxing constraints on GTF format)
-
Added new tests to overlapping.
-
Excluding regions in ologram is now done in C++, bringing major speedups (400x).
-
Added -b/--bool argument to overlapping.
-
Added -@/--annotate-all to overlapping.
-
Added bigwig_to_bed.
-
Added --bed-incl argument to ologram. (fix #73)
-
The --dpi has been removed from ologram.
-
The --pval-precision has been removed from ologram.
-
The --user-img-file argument in ologram has become pdf_file.
-
Added -W as command-wise argument. It force gtftk to print its message to a file. This may be handy when no access to stderr is available (e.g. through a scheduler).
This version provides several improvements and bug fix to ologram. The CLI of mk_matrix and join_multi_file have slightly changed. An int ([0-4]) is now mandatory to control verbosity level.
-
Fix #76 (issue with chromosomes in ensembl format when using ologram).
-
Fix #72 (P-values equal to zero when nb_intersections_true and nb_intersections_expectation_shuffled are equal to zero).
- Improved p-value precision in ologram.
-
Added -j/--sort-features to ologram. Controls the feature sorting in barplot diagram. Changed doc accordingly.
-
The positional argument 'bigwiglist' in mk_matrix has been replaced by -y.
-
The verbosity argument now must take a value (0-4).
-
The positional argument in join_multi_file has been replaced by -m.
- Updated changelog
- Improved p-value precision in ologram.
-
Improved p-value precision in ologram.
-
Added mpmath as dependency.
-
Added tiny_real dataset.
-
Added -q argument (--quiet) to get_example.
- No more using readthedoc as the doc is becoming too long to process.
- Many typo detected and fixed.
-
Added tiny_real dataset.
-
Added -q argument (--quiet) to get_example.
- Fix --no-strandness in divergent.
- Many typo detected and fixed.
-
This version now integrates ologram (OverLap Of Genomic Regions Analysis using Monte Carlo). Ologram annotates peaks(in BED format) with region sets/features extracted from (i) GTF file features (e.g promoter, tts, gene body, UTR...) (ii) GTF file keys (e.g. gene_biotype, user defined keys...) (iii) or from a BED file.
-
The user can now use --chrom-info to provide the command with a file or a string. The string should be one of 'mm8', 'mm9', 'mm10', 'hg19', 'hg38', 'rn3' or 'rn4'. When a genome version is requested as a string, the conventional chromosomes are used (chrM is discarded together with alternative haplotypes, unlocalized regions...).
-
Argparse was part of the dependencies. However, argparse is part of Python 3. Thus, this caused pygtftk to come with an older version of argparse...
-
Fixed gene sorting in tss_dict to ensure reproducible result.
-
Fixed a problem with retrieve() when used from interpreter (#45).
-
The load_gtf() function (C API) no raise an error if a GFF3 is passed with .gtf extension.
-
Input BED file in bed3 format are now converted to bed6 automatically.
-
The select_by_numeric() function has been renamed eval_numeric()
-
It is now possible to use numpy array of booleans to index the GTF (i.e. using the indexing function).
-
the prepare_gffutils_db() function allows one to create a db for gffutils while selecting features and attributes.
-
The argformatter module was refactored. Development of FormattedFile(argparse.FileType) that test for file extension and content (at least for bed).
-
The BED conversion is now performed using the print_bed() C function .
-
Fix a critical bug in get_sequence that affected get_feat_seq and get_tx_seq.
-
Select_by_key now throw an error when no key/val are available.
-
No more function with mutable objects as default arguments.
-
Fix temporary file deletion.
-
Refactored arg_formatter by creating a single type (ranged_num) to test for numeric inputs.
-
Refactored all plugins so that there is no more reference to unused arguments (tmp_dir, verbosity...).
-
No more reference to PY2.
-
Added several test to get_tx_seq and get_feat_seq.
-
Added several script to manipulate fasta files (see 'tools' folder). For pygtftk dev.
-
Added 'extra_require' slot in setup().
-
The get-feature-seq program now relies on bedtools (not on internal C code). This may change in the future as a a more flexible C interface is available.
-
Added --list-bigwigs to profile (to display the content of a coverage file).
-
Added a novel dataset (mini_real_10M) derived from mini_real and containing 10 Mb of chr1.
-
The configuration directory now supports several subdirectories named based on a hash string computed from path to the gtftk program.
- Convergent and divergent return coordinate (e.g. dist_to_divergent) in integer not float...
- pygtftk is no more compatible with python 2. This decision aims at integrating last plotnine versions (starting from 0.5.1) that depends on matplotlib 3.0.0 which strictly depends on py3k.
-
Added an example for col_from_tab in presentation.rst
-
Added a dataset mini_real_coding_pot.tab.
-
Working on travis now.
-
gtftk configuration directory now contains several subdirectories whose names are computed based on gtftk program location.
-
Added a -d argument to gtftk program. This argument returns gtftk configuration directory.
-
All tests should be independent of the directory.
-
Added test to count_key_values.
-
Fixed a bug in add_attr_from_file. The arg has_header led to empty result.
-
Fixed a bug in select_by_reg_exp. The match method was used instead of search method...
-
Fixed an error in count_key_values.
- Several changes in doc to make it compliant with readthedoc.
- Several changes in setup.py and requirements_develops.txt to makes it compliant with Pypi and buildable with manylinux.
- Several changes in setup.py and requirements_develops.txt to makes it compliant with Pypi and buildable with manylinux.
-
User-defined colors are now applied when calling profile command (bug in 0.9.2).
-
Fixed test issues when using md5sum under Linux.
-
Lots of typo fixed (their must be lot remaining unfortunatly...).
-
pygtftk is now compatible with both python 2.7 and >=3.5.
-
libstdcxx is no more required in env.yaml.
-
'Makefile clean' also reset version.py and conf.py to repository version.
-
Improved parser loading.
-
5p_3p_coord has been renamed get_5p_3p_coords.
-
Added a pip requirement.txt file for developers.
-
Added a specific conda environment file for py2k.
-
The default conda environment targets Python 3.6.
-
Improve garbage collector behavior upon exit.
-
Added --system-info argument to gtftk.