An R package (compatible with SCope) to create generic .loom files and extend them with other data e.g.: SCENIC regulons, Seurat clusters and markers, ... The package can also be used to read data from .loom files.

• HDF5 >= 1.10.1

For Linux and MacOS machines, version 1.10.1 can be installed with this snippet:

curl -O https://support.hdfgroup.org/ftp/HDF5/releases/hdf5-1.10/hdf5-1.10.1/src/hdf5-1.10.1.tar
cd hdf5-1.10.1
./configure
make -j4
make check
make install

To install HDF5 on a Windows machine, please use the prebuilt binaries available at https://www.hdfgroup.org/downloads/hdf5/.

For other HDF5 releases, please visit https://support.hdfgroup.org/ftp/HDF5/releases/.

Installation should take less than one minute.

# install.packages("devtools")
devtools::install_github("aertslab/SCopeLoomR")

• See how to create SCope compatible .loom files and extract data: Quick Start Tutorial.
• See how to create SCope compatible .loom files from a Seurat analysis: From Seurat Tutorial

September, 2021

• Version 0.13.0
  • Add function and tutorial section to get SCope all annotations (newly added) of clusterings get_all_clustering_annotations
• Version 0.12.0
  • Add function and tutorial section to get SCope annotations of a clustering get_clustering_annotations

April 08, 2021

• Version 0.11.0
  • Add tutorial to generate loom from Seurat analysis

April 07, 2021

• Version 0.10.5
  • Fix bug issue with get_clusterings_with_name when 1 clustering present only
  • Fix bug when adding seurat clusters & annotations aertslab#33

February 04, 2021

• Version 0.10.4
  • Fix bug issue with add_embedding: aertslab#32

November 24, 2020

• Version 0.10.2
  • Fix bug issue with get_clusterings_with_name and make it more robust: aertslab#29
  • Add row.names when using get_clusterings: aertslab#28

October 02, 2020

• Version 0.10.1
  • Avoid opening loom in "w" mode aertslab#26

September 22, 2020

• Version 0.10.0
  • Unifying functions to get cluster markers with get_cluster_markers (given clustering), get_all_cluster_markers (all clusterings)
  • By default, loom files are now loaded in read-only mode (open_loom)
  • Some function were reformatted/renamed:
    • get_clusterings_withName to get_clusterings_with_name
    • get_cellAnnotation to get_cell_annotation
    • get_regulonAuc to get_regulons_AUC
    • get_regulonThresholds to get_regulon_thresholds
    • get_regulons

August 17, 2020

• Version 0.9.3
  • Minor changes

June 22, 2020

• Version 0.9.2
  • Fix bug aertslab#17

May 27, 2020

• Version 0.9.1
  • Fix bug when clustering are of class character aertslab#15

May 27, 2020

• Version 0.9.1
  • Fix bug when clustering are of class character #15

March, 2020

• Version 0.9.0

  • Add functionality to get embedding by name
  • Fix bug update hierarchy for loom v3 spec
  • Fix bug gmd_clusterings not found when running add_annotation_by_cluster_annotation_mapping_df

• Version 0.8.0

  • Add functionality to update the cluster descriptions of given clustering using a mapping table (cluster ID to annotation)
  • Add functionality to add annotation from a mapping table (cluster ID to annotation)

• Version 0.7.0

  • Add functionality to convert loom with v2 specification to v3 specification. This conversion is recommended if you're planning to use the crowd annotation system from SCope.

February, 2020

• Version 0.6.4

  • Fix bug build_loom not working if dgem is a data.frame

• Version 0.6.3

  • Fix bug opening loom file v2
  • Fix bug using add_hierarchy results in broken loom file for SCope
  • Fix conditional statement in get_dspace

• Version 0.6.2

  • Fix bug datasets should not use scalar space
  • Fix bug global_meta_data_exists to work for loom v3 specs
  • Fix bug conditional statement to create attrs for loom v3 specs
  • Fix bug use json value in init_global_meta_data

• Version 0.6.1

  • Fix bug LOOM_SPEC_VERSION attribute does not exist for old Loom files generated with SCopeLoomR version < 0.6.0.

• Version 0.6.0

  • Add compatibility for Loom v3
  • Fix bug getting default embedding

January, 2020

• Version 0.5.1
  • Fix bug when add_seurat_clustering with annotation argument

June, 2019

• Version 0.5.0
  • Adding Seurat clustering results through add_seurat_clustering function works now also with Seurat v3 objects.

February, 2019

• Version 0.4.0
  • Embeddings and trajectories inferred from TI methods available within the dyno framework can be stored in .loom files and be displayed in SCope.

January, 2019

• Version 0.3.5

  • close_loom: New function
  • open_loom: Added argument "mode" and option to open as read-only (mode="r")

• Version 0.3.4

  • New functions to read SCENIC results (get_regulonsAuc, get_embeddings, get_regulons, get_cellAnnotation, get_clusterings_withName, get_regulonThresholds).
  • Updates to documentation.

January 11, 2019

• Version 0.3.3
  • Fix bug when adding Seurat clustering.

January 10, 2019

• Version 0.3.2
  • Fix bugs related to MetaData global attribute (absent or not compressed).
  • Add feature to overwrite default clustering if already existing one in the .loom.

November 8, 2018

• Version 0.3.1

  • Fix bug when adding sparse matrices.

• Version 0.3.0

  • Add feature to extract the gene expression matrix of a given cluster.
  • Update tutorial.

October 31, 2018

• Version 0.2.2
  • Add sanity tests when adding embeddings and SCENIC regulons with threshold assignments.
  • Small bug fixes.

August 8, 2018

• Version 0.2.1
  • Fix encoding bug leading to annotation not shown in SCope.
  • Add sanity test to check that all cells from the dgem are present in the Seurat clusters.

July 6, 2018

• Version 0.2.0
  • Add feature to store trajectory data. Currently, the following softwares can have their trajectory data easily stored in .loom files: monocle.

July 4, 2018

• Version 0.1.0
  • Add feature to store metrics