Skip to content

Version 1.1.8

Choose a tag to compare

View all tags
@patzaw patzaw released this 30 Jul 08:55
v1.1.8
1d371d2

Version 1.1.8 - Release notes - 2019-07-29

Implementation changes

  • Correction of how Ensembl data files are pre-processed because of
    the apparition of "carriage return" (\r\n) characters.
  • Give write rights on the neo4j database before copying it in the docker
    image.

Rebuild-BED.Rmd

  • Based on neo4j-community-3.5.7
  • Ensembl 97

Version 1.1.7 - Release notes - 2019-05-08

Implementation changes

  • When associating one id to another, the BE identified by the first id
    is deleted after that its production edges have been transferred.
    After this operation all id "corresponding_to" the first id do not
    directly identify any BE as they should be supposed to do. Thus,
    it's not recommended to associate such ID to others.

Version 1.1.6 - Release notes - 2019-04-23

Implementation changes

  • Split the toImport file in the bedImport function before loading csv
    in neo4j ==> improve robustness and efficiency of the load
  • Separate import of nodes and import of edges to avoid neo4j eager issues
  • Add Sus scrofa support in the dumpUniprotDb function
  • Ensembl history has been adapted to avoid loop (found for some pig
    transcripts)

Rebuild-BED.Rmd

  • Based on neo4j-community-3.5.4
  • Ensembl 96
  • Add GPL6887 platform
  • Add Sus scrofa (pig) support
  • Set the following cypher parameters during the building to avoid query
    replanning (https://neo4j.com/blog/cypher-write-fast-furious/):
    • --env=NEO4J_cypher_min__replan__interval=100000000ms \
    • --env=NEO4J_cypher_statistics__divergence__threshold=1 \
Assets 2
Loading