f742a96 | Daniel Himmelstein | 02 March 2017, 23:57:34 UTC | Upgrade to v0.2.2 | 02 March 2017, 23:57:34 UTC |
cbfd4c3 | Daniel Himmelstein | 02 March 2017, 23:42:23 UTC | Travis PyPI deploy: add wheel | 02 March 2017, 23:46:45 UTC |
2587193 | Daniel Himmelstein | 02 March 2017, 23:34:01 UTC | Limit deployment to Python 3.6 job only | 02 March 2017, 23:34:01 UTC |
a34946f | Daniel Himmelstein | 02 March 2017, 23:29:01 UTC | Remove register.py, switching to Travis deploy | 02 March 2017, 23:29:01 UTC |
8858d78 | Daniel Himmelstein | 02 March 2017, 23:19:48 UTC | Travis CI: deploy to PyPI | 02 March 2017, 23:23:09 UTC |
0b7baab | Daniel Himmelstein | 02 March 2017, 23:07:00 UTC | Travis CI: deploy to GitHub Releases | 02 March 2017, 23:23:09 UTC |
8f74660 | Daniel Himmelstein | 02 March 2017, 22:43:38 UTC | Use pandoc to generate README.rst Closes https://github.com/dhimmel/hetio/issues/7 | 02 March 2017, 22:47:41 UTC |
91562f7 | Daniel Himmelstein | 02 March 2017, 22:10:21 UTC | Expand setup.py lists | 02 March 2017, 22:10:21 UTC |
d026d13 | Tong Shu Li | 02 March 2017, 21:39:57 UTC | Change neo4j import queue size defaults Intended to speed up the hetnet importation into Neo4j. See https://github.com/dhimmel/hetio/issues/5 | 02 March 2017, 21:39:57 UTC |
bf8f548 | Daniel Himmelstein | 02 March 2017, 21:36:53 UTC | Minor cleaning of neo4j.py | 02 March 2017, 21:36:53 UTC |
7fe01f1 | Tong Shu Li | 02 March 2017, 21:28:39 UTC | Add py2neo version 3 URI compatibility (#8) Closes https://github.com/dhimmel/hetio/issues/4 Closes https://github.com/dhimmel/hetio/pull/8 Closes https://github.com/dhimmel/hetio/pull/6 | 02 March 2017, 21:28:39 UTC |
48b65b0 | Daniel Himmelstein | 16 February 2017, 00:28:25 UTC | Disable failing pytest feature in travis py34 | 16 February 2017, 00:28:25 UTC |
d32ba2a | Daniel Himmelstein | 16 February 2017, 00:23:24 UTC | Add support for bidirectional self-loop edges Do not allow the adding of duplicate edges. | 16 February 2017, 00:23:24 UTC |
8531e2b | Daniel Himmelstein | 15 February 2017, 23:35:03 UTC | Improve tests | 15 February 2017, 23:35:03 UTC |
7472d90 | Daniel Himmelstein | 15 February 2017, 23:06:00 UTC | Make abbreviation tests pytest discoverable | 15 February 2017, 23:06:00 UTC |
0799fe3 | Daniel Himmelstein | 15 February 2017, 22:28:42 UTC | Add Travis CI | 15 February 2017, 22:35:59 UTC |
903a4c8 | Daniel Himmelstein | 15 February 2017, 22:18:32 UTC | Count nodes and edges in graph | 15 February 2017, 22:18:32 UTC |
69aea62 | Daniel Himmelstein | 15 February 2017, 22:17:16 UTC | Fix tests with updated import path | 15 February 2017, 22:17:16 UTC |
042ba59 | Daniel Himmelstein | 29 September 2016, 00:48:20 UTC | Add pip install from PyPi command | 29 September 2016, 00:48:20 UTC |
321fc74 | Daniel Himmelstein | 29 September 2016, 00:47:11 UTC | Move pypandoc to dev but not install dependency Upload to PyPI as v0.2.1. register.py performs upload | 29 September 2016, 00:47:11 UTC |
ace55ab | Daniel Himmelstein | 05 September 2016, 19:45:59 UTC | Prepare 0.2.0 release and ready for PyPI | 05 September 2016, 19:45:59 UTC |
63a3737 | Daniel Himmelstein | 05 September 2016, 17:23:41 UTC | Add get_unicode_str function to BasePath | 05 September 2016, 17:23:41 UTC |
6d4a6f9 | Daniel Himmelstein | 20 May 2016, 18:22:39 UTC | Check type for edge/path equality | 20 May 2016, 18:22:39 UTC |
2c135fc | Daniel Himmelstein | 30 March 2016, 18:15:03 UTC | Add unicode metaedge str and style dataframe | 30 March 2016, 18:15:03 UTC |
7542a0d | Daniel Himmelstein | 22 March 2016, 05:28:33 UTC | Fix empty labeled unique node constraint bug | 22 March 2016, 05:28:33 UTC |
4a492cd | Daniel Himmelstein | 21 March 2016, 20:32:25 UTC | Separate index and join hints | 21 March 2016, 20:32:25 UTC |
9517fd4 | Daniel Himmelstein | 18 March 2016, 05:11:26 UTC | Fix non-xlsxwriter engine error | 18 March 2016, 05:11:26 UTC |
7ecb8c9 | Daniel Himmelstein | 18 March 2016, 00:44:11 UTC | Fix metanodes with no nodes bug | 18 March 2016, 00:44:11 UTC |
f7b1f67 | Daniel Himmelstein | 18 March 2016, 00:08:04 UTC | Freeze excel header rows | 18 March 2016, 00:08:04 UTC |
9e29b0c | Daniel Himmelstein | 17 March 2016, 18:30:59 UTC | Improve DWPC query Change `paths` to `path`. Singular better expresses the query plan. | 17 March 2016, 18:31:31 UTC |
e8bcd98 | Daniel Himmelstein | 17 March 2016, 18:24:53 UTC | Activate `unique_nodes` by default Simplify the `format_expanded_clause` query to use not equals (`<>`). | 17 March 2016, 18:24:53 UTC |
d16f26b | Daniel Himmelstein | 11 March 2016, 23:46:44 UTC | Merge branch 'master' of github.com:dhimmel/hetio | 11 March 2016, 23:46:44 UTC |
d524498 | Daniel Himmelstein | 11 March 2016, 23:46:34 UTC | Sort metanodes and metaeges in writable_from_metagraph | 11 March 2016, 23:46:34 UTC |
b53848c | Daniel Himmelstein | 08 March 2016, 15:20:07 UTC | Enable ftp:// and ftps:// url support | 08 March 2016, 15:20:07 UTC |
ac3d331 | Daniel Himmelstein | 29 February 2016, 00:49:37 UTC | Bug fixes | 29 February 2016, 00:49:37 UTC |
cdf7dfc | Daniel Himmelstein | 29 February 2016, 00:06:43 UTC | Add metanode to degree_df functionality | 29 February 2016, 00:06:43 UTC |
fe66a5b | Daniel Himmelstein | 28 February 2016, 22:40:05 UTC | Add join index hint Adds `USING JOIN ON` index hint in `construct_dwpc_query()` when `using=True`. Previously, this hint was not included due to a bug in neo4j 2.3.1. See https://github.com/neo4j/neo4j/pull/5038#issuecomment-162098241 | 28 February 2016, 22:40:05 UTC |
a6c8a28 | Daniel Himmelstein | 22 February 2016, 21:45:13 UTC | Standardize metaedge abbreviations Switch from arranging by text to non-inversion when standardizing metaedge abbreviations. Remove direction indicators (inequality symbols), in standardized abbreviations. | 22 February 2016, 21:45:13 UTC |
9681f11 | Daniel Himmelstein | 22 February 2016, 21:09:04 UTC | Exclude inverted edges The user will almost always want to see each metaedge once. Fixes a bug where undirected edges with the same source and target metanode had incorrect counts. | 22 February 2016, 21:09:04 UTC |
5e3647a | Daniel Himmelstein | 18 February 2016, 01:14:22 UTC | Add bzip2 read/write support Refactor file handling. | 18 February 2016, 01:14:22 UTC |
9370d71 | Daniel Himmelstein | 17 February 2016, 02:35:19 UTC | Change statistics reporting Switch from printing to a mix of logging and returning stats. Fixed a index bug in stat reporting. | 17 February 2016, 02:35:19 UTC |
9facc4b | Daniel Himmelstein | 06 January 2016, 16:48:18 UTC | Neo4j XSwap bug fixes Fix swap so: ``` u-->v y v ^ ^ becomes | | x-->y u x ``` Remove property transfer which was causing a relationship not found bug. Use cypher `id` function rather than attribute. Switch incorrect slashes to backslashes. Fix random index selection. | 06 January 2016, 17:00:44 UTC |
f8d23c3 | Daniel Himmelstein | 29 December 2015, 03:07:23 UTC | Add untested Xswap permutation | 29 December 2015, 03:07:23 UTC |
1aa1964 | Daniel Himmelstein | 19 December 2015, 23:30:19 UTC | Add permutation module To perform randomized edge swaps on the hetnet. | 19 December 2015, 23:30:19 UTC |
de3b607 | Daniel Himmelstein | 14 December 2015, 01:19:12 UTC | Make all Cypher functions lowercase | 14 December 2015, 01:19:12 UTC |
0683f5f | Daniel Himmelstein | 14 December 2015, 00:48:58 UTC | Label-aware expanded node uniqueness constraint Add an method for specifying node uniqueness in a path that leverages known node labels. See https://doi.org/10.15363/thinklab.d134#2 | 14 December 2015, 00:48:58 UTC |
f6deae3 | Daniel Himmelstein | 09 December 2015, 22:59:09 UTC | Allow unique node constraint Excluding paths with duplicate nodes is a potentially useful feature. See http://doi.org/10.15363/thinklab.d134 This commit adds an optional duplicate node exclusion for DWPC cypher queries. Thanks @ikwattro for suggesting the nested method. See https://twitter.com/ikwattro/status/674490872759631872. Thanks @antoine-lizee for suggesting the expanded method. See https://twitter.com/dhimmel/status/674487386949087232 | 09 December 2015, 22:59:09 UTC |
079edb2 | Daniel Himmelstein | 26 November 2015, 18:38:22 UTC | Add USING clauses to DWPC query Add an option (turned on by default) to include USING clauses in DWPC cypher queries. The USING option directs neo4j to traverse from both ends (http://neo4j.com/docs/2.3.1/query-using.html). The meet in the middle approach can lead to large efficiency improvements. See https://twitter.com/dhimmel/status/669933541514326016 for an example. Thanks @jexp for the input. | 26 November 2015, 18:38:22 UTC |
98121d6 | Daniel Himmelstein | 25 November 2015, 02:07:58 UTC | Fix metaedge to metarel conversion bug | 25 November 2015, 02:07:58 UTC |
d4b5f4e | Daniel Himmelstein | 25 November 2015, 01:00:08 UTC | Unique rel_type per metaedge Append abbreviation to rel_type so every metaedge has a unique rel_type. Previous implementation only appended the abbreviation for metaedge to metarel conversion. Add caching for metanode to label conversion, metaedge to rel_type conversion, and metaedge to metarel conversion. | 25 November 2015, 01:00:08 UTC |
89502d8 | Daniel Himmelstein | 24 November 2015, 21:53:53 UTC | Update for pandas 17 | 24 November 2015, 21:53:53 UTC |
a59c85f | Daniel Himmelstein | 24 November 2015, 19:35:18 UTC | Improve readability Add some docstrings Reorder code blocks Alphabetically order imports | 24 November 2015, 19:35:18 UTC |
908f183 | Daniel Himmelstein | 24 November 2015, 19:24:11 UTC | Remove ordered option for writable_from_graph() OrderedDict to Dict conversion is easy, so no need to explicitly support creating Dict writables. Give yaml.dump() a Dict rather than an OrderedDict to avoid error. | 24 November 2015, 19:24:11 UTC |
c9dfafd | Daniel Himmelstein | 24 November 2015, 16:30:07 UTC | Remove GML capability The GML is poorly documented. We haven't tested our support in a while and haven't used the feature. Removed for safety. | 24 November 2015, 16:30:07 UTC |
4874a0e | Daniel Himmelstein | 24 November 2015, 16:28:40 UTC | Auto detect format and support metagraph io | 24 November 2015, 16:28:40 UTC |
446b959 | Daniel Himmelstein | 24 November 2015, 00:22:19 UTC | Enable url reading | 24 November 2015, 00:22:19 UTC |
306ddb1 | Daniel Himmelstein | 23 November 2015, 23:53:09 UTC | Add metagraph files for testing | 23 November 2015, 23:53:09 UTC |
61d831f | Daniel Himmelstein | 06 November 2015, 16:47:37 UTC | Remove object reference in metagraph writable | 06 November 2015, 16:47:37 UTC |
49b22aa | Daniel Himmelstein | 06 November 2015, 16:37:34 UTC | Separate metagraph handling for writables Enables creating a writable of only the metagraph or creating a metagraph only from a writable. | 06 November 2015, 16:37:34 UTC |
8107d78 | Daniel Himmelstein | 06 November 2015, 02:41:14 UTC | Make rel_types unique for all metaedges @antoine-lizee discovered excessive node labeling in our Cypher queries was causing performance nightmares. See https://twitter.com/dhimmel/status/662440818398007296 Therefore, switch to neo4j graphs where each metaedge produces a different relationship type. Accomplish by append the metaedge abbreviation to the end of the rel type. For example, change `ASSOCIATION` to `ASSOCIATION_GaD` which specifies that it's a gene-disease association. Metaedge abbreviations are already guaranteed to be unique (see `abbreviation.validate_abbreviations()`). Remove source, target, and w from being returning by DWPC queries because the user will have these values. Remove intermediary node label declaration in path queries. Remove label declarations in DWPC path degree computation. | 06 November 2015, 02:41:14 UTC |
dc9c6eb | Daniel Himmelstein | 30 October 2015, 22:32:22 UTC | Move yaml import into functions | 30 October 2015, 22:32:22 UTC |
f3bcd7a | Daniel Himmelstein | 27 October 2015, 02:05:06 UTC | Return zero DWPC when no matches | 27 October 2015, 02:05:06 UTC |
53d1b17 | Daniel Himmelstein | 20 October 2015, 17:38:08 UTC | Return additional items in DWPC query | 20 October 2015, 17:38:08 UTC |
1582775 | Daniel Himmelstein | 20 October 2015, 17:14:17 UTC | Allow node property specification for cypher Added the property argument to `construct_dwpc_query()` to allow specification of which property to match the source and target nodes based on. Switched from using LENGTH to SIZE for calculating path degrees in cypher. Change suggested by @nicolewhite on twitter: https://twitter.com/_nicolemargaret/status/655165457079517184 | 20 October 2015, 17:14:17 UTC |
3150399 | Daniel Himmelstein | 16 October 2015, 20:27:44 UTC | Assign variable name to cypher DWPC output | 16 October 2015, 20:27:44 UTC |
1fe68cf | Daniel Himmelstein | 16 October 2015, 20:25:41 UTC | Fix bug when in metapath abbrev | 16 October 2015, 20:25:41 UTC |
a372b9c | Daniel Himmelstein | 14 October 2015, 21:29:12 UTC | Fix repr format bug | 14 October 2015, 21:29:12 UTC |
28711e8 | Daniel Himmelstein | 14 October 2015, 04:45:57 UTC | Bug fix | 14 October 2015, 04:45:57 UTC |
de6b3ad | Daniel Himmelstein | 14 October 2015, 04:24:53 UTC | Enable metapath retrieval from abbreviation | 14 October 2015, 04:24:53 UTC |
d76b017 | Daniel Himmelstein | 14 October 2015, 02:56:22 UTC | Add cypher query construction Construct metapath cypher query and DWPC cypher query. | 14 October 2015, 02:56:22 UTC |
1860faa | Daniel Himmelstein | 04 October 2015, 06:02:58 UTC | Add neo4j export | 04 October 2015, 06:02:58 UTC |
051ef5e | Daniel Himmelstein | 02 October 2015, 21:52:44 UTC | Update README.md Added Zenodo DOI badge | 02 October 2015, 21:52:44 UTC |
0a5b715 | Daniel Himmelstein | 17 September 2015, 00:04:57 UTC | Update README.md Update install command | 17 September 2015, 00:04:57 UTC |
75b7eb2 | Daniel Himmelstein | 16 September 2015, 21:56:56 UTC | Update README.md Added installation instructions | 16 September 2015, 21:56:56 UTC |
a006a86 | Daniel Himmelstein | 16 September 2015, 17:59:25 UTC | Update README.md Improved background and introduction | 16 September 2015, 17:59:25 UTC |
95a284f | Daniel Himmelstein | 02 September 2015, 19:26:07 UTC | Remove path extraction methods | 02 September 2015, 19:26:07 UTC |
d0357b0 | Daniel Himmelstein | 01 September 2015, 22:49:02 UTC | Using built-in types for dataframe Sorting was failing on pandas dataframes because order was not defined for nodes, metanodes, and edges. Switched so these elements are converted to strings before creation of the dataframe. Added a node_name column in the `get_degrees_for_metanode` output. | 01 September 2015, 22:49:02 UTC |
fc86dcc | Daniel Himmelstein | 01 September 2015, 21:45:23 UTC | Added dataframe sorting | 01 September 2015, 21:45:23 UTC |
101e773 | Daniel Himmelstein | 01 September 2015, 21:27:10 UTC | Abbreviation updates and testing | 01 September 2015, 21:27:10 UTC |
1042da3 | Daniel Himmelstein | 27 August 2015, 18:54:15 UTC | Updated MetaGraph.extract_metapaths Updated MetaGraph.extract_metapaths() to return an empty list when max_length is 0. Also added an assertion that max_length is not negative. | 27 August 2015, 18:54:15 UTC |
5e2879d | Daniel Himmelstein | 26 August 2015, 23:38:31 UTC | Added pathtools module Added pathtools module for calculating path counts and DWPCs. Modified metapath abbreviation to fit new abbreviation style. | 26 August 2015, 23:38:31 UTC |
2652cda | Daniel Himmelstein | 26 August 2015, 22:31:30 UTC | Update .gitignore | 26 August 2015, 22:31:30 UTC |
94206c9 | Daniel Himmelstein | 26 August 2015, 22:29:18 UTC | Merge branch 'master' of github.com:dhimmel/hetio | 26 August 2015, 22:29:18 UTC |
869a678 | Daniel Himmelstein | 26 August 2015, 22:27:03 UTC | Renamed module and moved tests Renamed graph module to hetnet. See https://dx.doi.org/10.15363/thinklab.d104 Moved tests to a dedicated directory. | 26 August 2015, 22:27:03 UTC |
403fd48 | Daniel Himmelstein | 26 August 2015, 21:54:51 UTC | Update LICENSE.md Switch from MIT to CC0 licensing | 26 August 2015, 21:54:51 UTC |
ddf4bba | Daniel Himmelstein | 14 August 2015, 17:30:01 UTC | Differentiated directional metaedge abbreviations Additionally added new stats methods for summarizing metanodes and metaedges. | 14 August 2015, 17:30:01 UTC |
20e4a48 | Daniel Himmelstein | 13 August 2015, 23:26:41 UTC | Added abbreviation module Added a validate_abbreviation function to check that abbreviations are valid and unambiguous. Added automatic abbreviation generation. The two previous additions close #1. Switched relative imports to absolution imports, since "Absolute imports are recommended, as they are usually more readable and tend to be better behaved" https://www.python.org/dev/peps/pep-0008/ | 13 August 2015, 23:26:41 UTC |
ef0b76b | Daniel Himmelstein | 13 August 2015, 00:33:02 UTC | Changes to abbreviations Abbreviations are now saved with a graph. For graphs saved without abbreviation information, the abbreviations can be automatically generated. However, this feature is not currently implemented, so saved graphs without `kind_to_abbrev` with raise a NotImplementedError. Automatic abbreviation generation code has been removed from graph.MetaGraph() because it was complex/lengthy and not the focus of the graph module. Currently, kind_to_abbrev is not checked for safety -- i.e. the user could provide ambiguous abbreviations without an error. | 13 August 2015, 00:33:02 UTC |
1f286f6 | Daniel Himmelstein | 11 August 2015, 02:17:34 UTC | Created network stats and visualization module Currently all functionality is related to extracting and visualizing node degree distributions. | 11 August 2015, 02:17:34 UTC |
1de7a01 | Daniel Himmelstein | 11 August 2015, 00:07:08 UTC | Added masked option to write_json() | 11 August 2015, 00:07:08 UTC |
4e260c4 | Daniel Himmelstein | 10 August 2015, 01:31:21 UTC | Added read_json method | 10 August 2015, 01:31:21 UTC |
340b5f3 | Daniel Himmelstein | 15 July 2015, 15:38:14 UTC | Fixed pernicious write_sif newline bug. | 15 July 2015, 15:38:14 UTC |
56f2839 | Daniel Himmelstein | 15 July 2015, 03:18:09 UTC | Update README.md Added project hyperlinks and additional background. | 15 July 2015, 03:18:09 UTC |
4b842f3 | Daniel Himmelstein | 15 July 2015, 03:05:15 UTC | Enabled JSON encoding of numpy dtypes. Adjustements to new node attributes. | 15 July 2015, 03:05:15 UTC |
5cccc90 | Daniel Himmelstein | 15 July 2015, 02:03:48 UTC | Updated .sif writing and changed node __str__ to include metanode. | 15 July 2015, 02:03:48 UTC |
9c7a907 | Daniel Himmelstein | 02 July 2015, 23:08:06 UTC | Fixed a bug in Edge.get_id(). Began adding __str__ methods to work with the new node.name attribute | 02 July 2015, 23:08:06 UTC |
daa4c14 | Daniel Himmelstein | 01 July 2015, 21:07:45 UTC | Updates for (metanode, node) identification support. | 01 July 2015, 21:07:45 UTC |
3c06dfd | Daniel Himmelstein | 01 July 2015, 21:00:47 UTC | Added install script | 01 July 2015, 21:00:47 UTC |
d769ca2 | Daniel Himmelstein | 01 July 2015, 01:26:26 UTC | Reordered Node() and Graph.add_node() arguments and created a Node attribute called name for display purposes | 01 July 2015, 01:26:26 UTC |
6b03470 | Daniel Himmelstein | 01 July 2015, 01:10:03 UTC | Added minimal testing using pytest. | 01 July 2015, 01:10:03 UTC |