Skip to content
Snippets Groups Projects
Select Git revision
  • 4-add-test-ci
  • inferences
  • main default protected
  • package
  • refactoring
  • 1.0.3
  • 1.0.2
  • 1.0.1
  • 1.0.0
9 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.012Mar1154326Feb30Jan2924211528Nov20191824Oct18713Sep114229Aug222023Jul1213Jun126329May2824222117151324Apr2322181210843222Mar2113121185429Feb282321201915128526Jan25242218121110947Dec5410Nov76331Oct3026242320161210654225Jul208Jun26May252217521Apr20changed feature_on_target_genome occurence into a class FeaturePathmainmainconverted segments_on_target_genome to use a new class SegmentTarget; clarifies how the dictionnary is usedfused the dictionnaries seg_len/seg_size and seg_seq into seg_info that contains both the size and the sequence of the segments.made feature id detection less dependant on the gff structurefixed errors between feature key and feature id; this fixes a bug when a feature is split on several lines int he source gffmoved functions in intersect.py for clarity; moved more functions into class Featureadded a few functions to the class Feature; replaced the parent/child feature_id attribute by the corresponding class instancemodified how the parent field is detected in gff, made it more flexibleturn feature path list into set to check faster if seg is present, in var detail and aln; ignore case where a segment is twice in a feature pathadded dataverse linkremoved yates continuity correctionadded scripts for publicationadded the feature strand in target genome in variation details output fileruntime opti, changed feature_seg_list in dictionnary to check if seg present in list, to compute child features paths and to compute cov/id; much faster on human data, not so much on rice and ecoli datafix bug when there are several haplotypes and option -ht is used; didnt fix bug when there are several haplotypes but option -ht is absent, todo add a warningMerge branch 'refactoring'improved how paths and walks are detected and how bed files are attributed to genomes; fixes bug when a genome name was included in another genome namerefactoringrefactoringbug fix when a segment is present in both orientation in the source features; modified search_segment to find the orientation that corresponds to the current featurecorrected feature info print in class Featurebug fix; with several chromosomes the features were transfered only on the last one loadedremove buggy releaseMerge branch '3-add-ci' into 'main'4-add-test-ci4-add-test-ciUpdate .gitlab-ci.yml file -TESTUpdate .gitlab-ci.yml file -TEST-TEST-TESTTEST add cimoved functions into the class as methods; added a few attributesfinished sorting the functions in several filesemptied this file by putting the functions in other filesrenamed filecontinued code refactoring. removed global variable segments_on_target_genome, sorted functions in several filesattempt to reorganize the python fileslimited imports between filessplit functions from Graph_gff in two files, load_intersect and graph_annotadded option to not output duplicated features in annotation, only the best occurence is kept. corrected the alignment option, it now doesnt print features that didnt pass the filters.added new version for benchmarkload intersect is now much quicker. however there might be unexpected behaviour in detect_gene_copies (or maybe elsewhere) if a segment is present several times in a feature (duplication within the feature). this case was not encountered in rice and human data, but it is technically possible.Merge branch 'main' of forge.ird.fr:diade/dynadiv/grannotdelete the intersect file after loading it, it takes too much space