hAMRonization 1.3.1__tar.gz → 1.4.1__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (40) hide show
  1. {hamronization-1.3.1 → hamronization-1.4.1}/PKG-INFO +79 -12
  2. {hamronization-1.3.1 → hamronization-1.4.1}/README.md +77 -10
  3. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/AmrFinderPlusIO.py +58 -13
  4. hamronization-1.4.1/hAMRonization/DatabaseMapping.py +453 -0
  5. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/Interfaces.py +53 -3
  6. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/RgiIO.py +6 -0
  7. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/TBProfilerIO.py +22 -9
  8. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/__init__.py +1 -1
  9. hamronization-1.4.1/hAMRonization/data_sources.py +90 -0
  10. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization.egg-info/PKG-INFO +79 -12
  11. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization.egg-info/SOURCES.txt +4 -1
  12. {hamronization-1.3.1 → hamronization-1.4.1}/setup.cfg +1 -1
  13. {hamronization-1.3.1 → hamronization-1.4.1}/setup.py +26 -2
  14. hamronization-1.4.1/test/test_db_mapping.py +249 -0
  15. {hamronization-1.3.1 → hamronization-1.4.1}/test/test_parsing_validity.py +84 -0
  16. {hamronization-1.3.1 → hamronization-1.4.1}/LICENSE.txt +0 -0
  17. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/AbricateIO.py +0 -0
  18. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/AmrPlusPlusIO.py +0 -0
  19. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/AribaIO.py +0 -0
  20. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/CSStarIO.py +0 -0
  21. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/DeepArgIO.py +0 -0
  22. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/FARGeneIO.py +0 -0
  23. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/GrootIO.py +0 -0
  24. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/KmerResistanceIO.py +0 -0
  25. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/MykrobeIO.py +0 -0
  26. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/ResFamsIO.py +0 -0
  27. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/ResFinderIO.py +0 -0
  28. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/SraxIO.py +0 -0
  29. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/Srst2IO.py +0 -0
  30. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/StarAmrIO.py +0 -0
  31. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/constants.py +0 -0
  32. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/hAMRonizedResult.py +0 -0
  33. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/hamronize.py +0 -0
  34. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization/summarize.py +0 -0
  35. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization.egg-info/dependency_links.txt +0 -0
  36. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization.egg-info/entry_points.txt +0 -0
  37. /hamronization-1.3.1/hAMRonization.egg-info/zip-safe → /hamronization-1.4.1/hAMRonization.egg-info/not-zip-safe +0 -0
  38. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization.egg-info/requires.txt +0 -0
  39. {hamronization-1.3.1 → hamronization-1.4.1}/hAMRonization.egg-info/top_level.txt +0 -0
  40. {hamronization-1.3.1 → hamronization-1.4.1}/test/test_interfaces.py +0 -0
@@ -1,9 +1,9 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: hAMRonization
3
- Version: 1.3.1
3
+ Version: 1.4.1
4
4
  Summary: Tool to convert and summarize AMR gene detection outputs using the hAMRonization specification
5
5
  Home-page: https://github.com/pha4ge/hAMRonization
6
- Download-URL: https://github.com/pha4ge/archive/v1.3.1.tar.gz
6
+ Download-URL: https://github.com/pha4ge/archive/v1.4.1.tar.gz
7
7
  Author: Finlay Maguire
8
8
  Author-email: finlaymaguire@gmail.com
9
9
  License: LGPLv3
@@ -83,8 +83,8 @@ conda create --name hamronization --channel conda-forge --channel bioconda hamro
83
83
 
84
84
  Or to install and run using docker, podman, singularity:
85
85
  ```
86
- docker pull docker.io/finlaymaguire/hamronization:latest
87
- docker run --rm docker.io/finlaymaguire/hamronization:latest hamronize --help
86
+ docker pull ghcr.io/pha4ge/hamronization:latest
87
+ docker run --rm ghcr.io/pha4ge/hamronization:latest hamronize --help
88
88
  ```
89
89
 
90
90
  Or to install the latest development version:
@@ -169,6 +169,50 @@ and they will be concatenated appropriately (i.e. only one header for tsv)
169
169
  hamronize rgi --input_file_name rgi_report --analysis_software_version 6.0.0 --reference_database_version 3.2.5 test/data/raw_outputs/rgi/rgi.txt test/data/raw_outputs/rgibwt/Kp11_bwtoutput.gene_mapping_data.txt
170
170
  ```
171
171
 
172
+ #### Mapping to other AMR databases (RGI, AMRFinderPlus, ResFinder)
173
+
174
+ The results of `rgi` (CARD), `amrfinderplus` (NCBI Reference Gene Database) and
175
+ `resfinder` (ResFinder) can optionally be mapped to the other two of these
176
+ databases with `--db_mapping chamredb` or `--db_mapping argnorm`:
177
+
178
+ ```
179
+ hamronize amrfinderplus --db_mapping chamredb --analysis_software_version 4.0.3 --reference_database_version 2024-07-22.1 --input_file_name sample1 amrfinderplus_report.tsv
180
+ ```
181
+
182
+ This appends a `<database>_mapping` column for each of the other two databases
183
+ (e.g. `card_mapping` and `resfinder_mapping` for AMRFinderPlus) and a
184
+ `database_mapping_source` column recording which resource was used. Multiple
185
+ mapped entries are separated by ` | `, and the columns are left empty where no
186
+ mapping is found.
187
+
188
+ - [chamredb](https://github.com/antunderwood/chamredb) uses a graph of
189
+ reciprocal best hits (RBH) between CARD, NCBI and ResFinder sequences. The
190
+ RBHs are reported for each database, or the one-way hits (OWH) where there is
191
+ no RBH, with their identity and coverage, e.g.
192
+ `ARO:3001022 (TEM-156; RBH; identity=1.0; coverage=1.0)`. Hits are looked up
193
+ by accession, then by gene name.
194
+ - [argNorm](https://github.com/BigDataBiology/argNorm) maps NCBI and ResFinder
195
+ genes onto CARD ARO terms, e.g. `ARO:3001022 (TEM-156; Perfect)`. Mappings
196
+ from CARD, and between NCBI and ResFinder, go via the ARO term, e.g.
197
+ `blaTEM-156 (blaTEM-156_1_AM941159; via ARO:3001022; Perfect)`. argNorm's
198
+ `Loose` matches are not used for these indirect mappings, and only the best
199
+ available cut-off is reported. Alleles sharing a gene name are grouped
200
+ (`+N more`).
201
+
202
+ Pinned copies of chamredb's
203
+ [`graph.json`](https://github.com/antunderwood/chamredb/blob/master/chamredb/data/graph.json)
204
+ (2026-11-09 graph) and argNorm's (v1.1.0)
205
+ [mapping tables](https://github.com/BigDataBiology/argNorm/tree/main/argnorm/data)
206
+ are downloaded, with their checksums verified, when hAMRonization is installed
207
+ (see `hAMRonization/data_sources.py`). This needs network access at install
208
+ time. Set `HAMRONIZATION_SKIP_DATA_DOWNLOAD=1` to install without them. To use
209
+ a different copy, pass `--db_mapping_data` with the path to a `graph.json` or
210
+ an argNorm `data` directory. Both resources were built against specific
211
+ database releases, so genes added more recently may not map.
212
+
213
+ When `--db_mapping` is used, `hamronize` prints what to cite to stderr (see
214
+ [Citation](#citation)).
215
+
172
216
  You can summarize hAMRonized reports regardless of format using the 'summarize'
173
217
  function:
174
218
 
@@ -225,6 +269,17 @@ You can also output a `json` formatted hAMRonized report:
225
269
 
226
270
  `parsed_report.write('all_hAMRonized_abricate_report.json', output_format='json')`
227
271
 
272
+ To add the database mapping columns (see above) when writing an `rgi`,
273
+ `amrfinderplus` or `resfinder` report, pass a mapper; `mapper.map(result)` also
274
+ returns the mapping for a single result as a dict:
275
+
276
+ ```
277
+ from hAMRonization.DatabaseMapping import get_mapper
278
+ mapper = get_mapper("argnorm", "rgi") # or get_mapper("chamredb", "rgi", "path/to/graph.json")
279
+ print(mapper.citation_message) # what to cite for this mapping
280
+ hAMRonization.parse("rgi.txt", metadata, "rgi").write(output_location="rgi.tsv", db_mapper=mapper)
281
+ ```
282
+
228
283
  If you want to write multiple reports to one file, this `.write` method can accept `append_mode=True` to append rather than overwrite the output file and not include the header (in tsv format).
229
284
 
230
285
  `parsed_report.write('all_hAMRonized_abricate_report.tsv', append_mode=True)`
@@ -234,24 +289,24 @@ If you want to write multiple reports to one file, this `.write` method can acce
234
289
 
235
290
  Currently implemented parsers and the last tool version for which they have been validated:
236
291
 
237
- 1. [abricate](hAMRonization/AbricateIO.py): last updated for v1.0.0
238
- 2. [amrfinderplus](hAMRonization/AmrFinderPlusIO.py): last updated for v4.0.3
239
- 3. [amrplusplus](hAMRonization/AmrPlusPlusIO.py): last updated for c6b097a
292
+ 1. [abricate](hAMRonization/AbricateIO.py): last updated for v1.4.0
293
+ 2. [amrfinderplus](hAMRonization/AmrFinderPlusIO.py): last updated for v4.2.7
294
+ 3. [amrplusplus](hAMRonization/AmrPlusPlusIO.py): last updated for v4 (MEGARes v4.00)
240
295
  4. [ariba](hAMRonization/AribaIO.py): last updated for v2.14.6
241
296
  5. [csstar](hAMRonization/CSStarIO.py): last updated for v2.1.0
242
- 6. [deeparg](hAMRonization/DeepArgIO.py): last updated for v1.0.2
297
+ 6. [deeparg](hAMRonization/DeepArgIO.py): last updated for v2.0.0
243
298
  7. [fargene](hAMRonization/FARGeneIO.py): last updated for v0.1
244
299
  8. [groot](hAMRonization/GrootIO.py): last updated for v1.1.2
245
300
  9. [kmerresistance](hAMRonization/KmerResistanceIO.py): late updated for v2.2.0
246
301
  10. [mykrobe](test/data/raw_outputs/mykrobe/report.json): last updated for v0.8.1
247
302
  11. ~pointfinder~ (removed, PointFinder is now integrated in ResFinder)
248
303
  12. [resfams](hAMRonization/ResFamsIO.py): last updated for hmmer v3.3.2
249
- 13. [resfinder](hAMRonization/ResFinderIO.py): last updated for v4.6.0
250
- 14. [rgi](hAMRonization/RgiIO.py) (includes RGI-BWT) last updated for v5.2.0
304
+ 13. [resfinder](hAMRonization/ResFinderIO.py): last updated for v4.7.3
305
+ 14. [rgi](hAMRonization/RgiIO.py) (includes RGI-BWT) last updated for v6.0.8
251
306
  15. [srax](hAMRonization/SraxIO.py): last updated for v1.5
252
307
  16. [srst2](hAMRonization/Srst2IO.py): last updated for v0.2.0
253
- 17. [staramr](hAMRonization/StarAmrIO.py): last updated for v0.8.0
254
- 18. [tbprofilder](test/data/raw_outputs/tbprofiler/tbprofiler.json): last updated for v3.0.8
308
+ 17. [staramr](hAMRonization/StarAmrIO.py): last updated for v0.12.3
309
+ 18. [tbprofiler](hAMRonization/TBProfilerIO.py): last updated for v7.0.0 (also parses v3/v4 output)
255
310
 
256
311
  ## Implementation Details
257
312
 
@@ -388,6 +443,18 @@ Here are some known issues that we would welcome input on trying to solve!
388
443
 
389
444
  - `contig_id` isn't general enough when some tools this ID naturally corresponds to a `read_name` (deepARG), individual ORF (resfams), or protein sequence (AFP with protein input): *change to `query_id_name` or similar?*
390
445
 
446
+
447
+ #### Limitations of cross-database mapping
448
+
449
+ This is a highly messy and approximate process so any cross-mappings should be taken with a major grain of salt and directly validated if being used for important purposes.
450
+
391
451
  ## Citation
392
452
  If you use hAMRonization please cite the following publication:
393
453
  > I Mendes et al. 2024. "hAMRonization: Enhancing antimicrobial resistance prediction using the PHA4GE AMR detection specification and tooling". bioRxiv. https://doi.org/10.1101/2024.03.07.583950
454
+
455
+ If you use `--db_mapping chamredb`, also reference [chamredb](https://github.com/antunderwood/chamredb). It has no separate publication or preprint at this point.
456
+
457
+ If you use `--db_mapping argnorm`, please also cite:
458
+ > S Ugarcina Perovic, V Ramji, H Chong, Y Duan, F Maguire, LP Coelho. 2025. "argNorm: normalization of antibiotic resistance gene annotations to the Antibiotic Resistance Ontology (ARO)". Bioinformatics. https://doi.org/10.1093/bioinformatics/btaf173
459
+
460
+ `hamronize` prints the relevant citations to stderr whenever `--db_mapping` is used.
@@ -39,8 +39,8 @@ conda create --name hamronization --channel conda-forge --channel bioconda hamro
39
39
 
40
40
  Or to install and run using docker, podman, singularity:
41
41
  ```
42
- docker pull docker.io/finlaymaguire/hamronization:latest
43
- docker run --rm docker.io/finlaymaguire/hamronization:latest hamronize --help
42
+ docker pull ghcr.io/pha4ge/hamronization:latest
43
+ docker run --rm ghcr.io/pha4ge/hamronization:latest hamronize --help
44
44
  ```
45
45
 
46
46
  Or to install the latest development version:
@@ -125,6 +125,50 @@ and they will be concatenated appropriately (i.e. only one header for tsv)
125
125
  hamronize rgi --input_file_name rgi_report --analysis_software_version 6.0.0 --reference_database_version 3.2.5 test/data/raw_outputs/rgi/rgi.txt test/data/raw_outputs/rgibwt/Kp11_bwtoutput.gene_mapping_data.txt
126
126
  ```
127
127
 
128
+ #### Mapping to other AMR databases (RGI, AMRFinderPlus, ResFinder)
129
+
130
+ The results of `rgi` (CARD), `amrfinderplus` (NCBI Reference Gene Database) and
131
+ `resfinder` (ResFinder) can optionally be mapped to the other two of these
132
+ databases with `--db_mapping chamredb` or `--db_mapping argnorm`:
133
+
134
+ ```
135
+ hamronize amrfinderplus --db_mapping chamredb --analysis_software_version 4.0.3 --reference_database_version 2024-07-22.1 --input_file_name sample1 amrfinderplus_report.tsv
136
+ ```
137
+
138
+ This appends a `<database>_mapping` column for each of the other two databases
139
+ (e.g. `card_mapping` and `resfinder_mapping` for AMRFinderPlus) and a
140
+ `database_mapping_source` column recording which resource was used. Multiple
141
+ mapped entries are separated by ` | `, and the columns are left empty where no
142
+ mapping is found.
143
+
144
+ - [chamredb](https://github.com/antunderwood/chamredb) uses a graph of
145
+ reciprocal best hits (RBH) between CARD, NCBI and ResFinder sequences. The
146
+ RBHs are reported for each database, or the one-way hits (OWH) where there is
147
+ no RBH, with their identity and coverage, e.g.
148
+ `ARO:3001022 (TEM-156; RBH; identity=1.0; coverage=1.0)`. Hits are looked up
149
+ by accession, then by gene name.
150
+ - [argNorm](https://github.com/BigDataBiology/argNorm) maps NCBI and ResFinder
151
+ genes onto CARD ARO terms, e.g. `ARO:3001022 (TEM-156; Perfect)`. Mappings
152
+ from CARD, and between NCBI and ResFinder, go via the ARO term, e.g.
153
+ `blaTEM-156 (blaTEM-156_1_AM941159; via ARO:3001022; Perfect)`. argNorm's
154
+ `Loose` matches are not used for these indirect mappings, and only the best
155
+ available cut-off is reported. Alleles sharing a gene name are grouped
156
+ (`+N more`).
157
+
158
+ Pinned copies of chamredb's
159
+ [`graph.json`](https://github.com/antunderwood/chamredb/blob/master/chamredb/data/graph.json)
160
+ (2026-11-09 graph) and argNorm's (v1.1.0)
161
+ [mapping tables](https://github.com/BigDataBiology/argNorm/tree/main/argnorm/data)
162
+ are downloaded, with their checksums verified, when hAMRonization is installed
163
+ (see `hAMRonization/data_sources.py`). This needs network access at install
164
+ time. Set `HAMRONIZATION_SKIP_DATA_DOWNLOAD=1` to install without them. To use
165
+ a different copy, pass `--db_mapping_data` with the path to a `graph.json` or
166
+ an argNorm `data` directory. Both resources were built against specific
167
+ database releases, so genes added more recently may not map.
168
+
169
+ When `--db_mapping` is used, `hamronize` prints what to cite to stderr (see
170
+ [Citation](#citation)).
171
+
128
172
  You can summarize hAMRonized reports regardless of format using the 'summarize'
129
173
  function:
130
174
 
@@ -181,6 +225,17 @@ You can also output a `json` formatted hAMRonized report:
181
225
 
182
226
  `parsed_report.write('all_hAMRonized_abricate_report.json', output_format='json')`
183
227
 
228
+ To add the database mapping columns (see above) when writing an `rgi`,
229
+ `amrfinderplus` or `resfinder` report, pass a mapper; `mapper.map(result)` also
230
+ returns the mapping for a single result as a dict:
231
+
232
+ ```
233
+ from hAMRonization.DatabaseMapping import get_mapper
234
+ mapper = get_mapper("argnorm", "rgi") # or get_mapper("chamredb", "rgi", "path/to/graph.json")
235
+ print(mapper.citation_message) # what to cite for this mapping
236
+ hAMRonization.parse("rgi.txt", metadata, "rgi").write(output_location="rgi.tsv", db_mapper=mapper)
237
+ ```
238
+
184
239
  If you want to write multiple reports to one file, this `.write` method can accept `append_mode=True` to append rather than overwrite the output file and not include the header (in tsv format).
185
240
 
186
241
  `parsed_report.write('all_hAMRonized_abricate_report.tsv', append_mode=True)`
@@ -190,24 +245,24 @@ If you want to write multiple reports to one file, this `.write` method can acce
190
245
 
191
246
  Currently implemented parsers and the last tool version for which they have been validated:
192
247
 
193
- 1. [abricate](hAMRonization/AbricateIO.py): last updated for v1.0.0
194
- 2. [amrfinderplus](hAMRonization/AmrFinderPlusIO.py): last updated for v4.0.3
195
- 3. [amrplusplus](hAMRonization/AmrPlusPlusIO.py): last updated for c6b097a
248
+ 1. [abricate](hAMRonization/AbricateIO.py): last updated for v1.4.0
249
+ 2. [amrfinderplus](hAMRonization/AmrFinderPlusIO.py): last updated for v4.2.7
250
+ 3. [amrplusplus](hAMRonization/AmrPlusPlusIO.py): last updated for v4 (MEGARes v4.00)
196
251
  4. [ariba](hAMRonization/AribaIO.py): last updated for v2.14.6
197
252
  5. [csstar](hAMRonization/CSStarIO.py): last updated for v2.1.0
198
- 6. [deeparg](hAMRonization/DeepArgIO.py): last updated for v1.0.2
253
+ 6. [deeparg](hAMRonization/DeepArgIO.py): last updated for v2.0.0
199
254
  7. [fargene](hAMRonization/FARGeneIO.py): last updated for v0.1
200
255
  8. [groot](hAMRonization/GrootIO.py): last updated for v1.1.2
201
256
  9. [kmerresistance](hAMRonization/KmerResistanceIO.py): late updated for v2.2.0
202
257
  10. [mykrobe](test/data/raw_outputs/mykrobe/report.json): last updated for v0.8.1
203
258
  11. ~pointfinder~ (removed, PointFinder is now integrated in ResFinder)
204
259
  12. [resfams](hAMRonization/ResFamsIO.py): last updated for hmmer v3.3.2
205
- 13. [resfinder](hAMRonization/ResFinderIO.py): last updated for v4.6.0
206
- 14. [rgi](hAMRonization/RgiIO.py) (includes RGI-BWT) last updated for v5.2.0
260
+ 13. [resfinder](hAMRonization/ResFinderIO.py): last updated for v4.7.3
261
+ 14. [rgi](hAMRonization/RgiIO.py) (includes RGI-BWT) last updated for v6.0.8
207
262
  15. [srax](hAMRonization/SraxIO.py): last updated for v1.5
208
263
  16. [srst2](hAMRonization/Srst2IO.py): last updated for v0.2.0
209
- 17. [staramr](hAMRonization/StarAmrIO.py): last updated for v0.8.0
210
- 18. [tbprofilder](test/data/raw_outputs/tbprofiler/tbprofiler.json): last updated for v3.0.8
264
+ 17. [staramr](hAMRonization/StarAmrIO.py): last updated for v0.12.3
265
+ 18. [tbprofiler](hAMRonization/TBProfilerIO.py): last updated for v7.0.0 (also parses v3/v4 output)
211
266
 
212
267
  ## Implementation Details
213
268
 
@@ -344,6 +399,18 @@ Here are some known issues that we would welcome input on trying to solve!
344
399
 
345
400
  - `contig_id` isn't general enough when some tools this ID naturally corresponds to a `read_name` (deepARG), individual ORF (resfams), or protein sequence (AFP with protein input): *change to `query_id_name` or similar?*
346
401
 
402
+
403
+ #### Limitations of cross-database mapping
404
+
405
+ This is a highly messy and approximate process so any cross-mappings should be taken with a major grain of salt and directly validated if being used for important purposes.
406
+
347
407
  ## Citation
348
408
  If you use hAMRonization please cite the following publication:
349
409
  > I Mendes et al. 2024. "hAMRonization: Enhancing antimicrobial resistance prediction using the PHA4GE AMR detection specification and tooling". bioRxiv. https://doi.org/10.1101/2024.03.07.583950
410
+
411
+ If you use `--db_mapping chamredb`, also reference [chamredb](https://github.com/antunderwood/chamredb). It has no separate publication or preprint at this point.
412
+
413
+ If you use `--db_mapping argnorm`, please also cite:
414
+ > S Ugarcina Perovic, V Ramji, H Chong, Y Duan, F Maguire, LP Coelho. 2025. "argNorm: normalization of antibiotic resistance gene annotations to the Antibiotic Resistance Ontology (ARO)". Bioinformatics. https://doi.org/10.1093/bioinformatics/btaf173
415
+
416
+ `hamronize` prints the relevant citations to stderr whenever `--db_mapping` is used.
@@ -16,6 +16,47 @@ required_metadata = [
16
16
  "input_file_name",
17
17
  ]
18
18
 
19
+ # Subtypes whose element symbol is <gene>_<mutation>. POINT_DISRUPT (v4.2.4+)
20
+ # are frameshifts/stop codons in genes whose disruption causes resistance.
21
+ POINT_SUBTYPES = {"POINT", "POINT_DISRUPT"}
22
+
23
+ # <gene>_<ref><pos>[_<end_ref><end_pos>]<alt>, e.g. gyrA_S83I, mtrR_G-131A,
24
+ # cirA_Y253CfsTer5, pmrB_ALNQLV129del, mexR_D34_F96del, folP_P64PSFLY, see
25
+ # https://github.com/ncbi/amr/wiki/Point-mutation-nomenclature
26
+ POINT_MUTATION_RE = re.compile(
27
+ r"^(?P<gene>.+?)_(?P<ref>[A-Za-z]+)(?P<pos>-?\d+)"
28
+ r"(?:_(?P<end_ref>[A-Za-z]+)(?P<end_pos>-?\d+))?"
29
+ r"(?P<alt>[A-Za-z][A-Za-z0-9]*)$"
30
+ )
31
+
32
+
33
+ def _shift(pos, offset):
34
+ """Offset a coordinate, skipping 0 (c.-1 is followed by c.1)"""
35
+ new = pos + offset
36
+ if pos < 0 <= new:
37
+ new += 1
38
+ return new
39
+
40
+
41
+ def _protein_mutation(m):
42
+ if m["end_pos"]:
43
+ return f"p.{m['ref']}{m['pos']}_{m['end_ref']}{m['end_pos']}{m['alt']}"
44
+ return f"p.{m['ref']}{m['pos']}{m['alt']}"
45
+
46
+
47
+ def _nucleotide_mutation(m):
48
+ ref, alt, pos = m["ref"], m["alt"], int(m["pos"])
49
+ end = int(m["end_pos"]) if m["end_pos"] else _shift(pos, len(ref) - 1)
50
+ span = f"{pos}" if end == pos else f"{pos}_{end}"
51
+ if alt == "del":
52
+ return f"c.{span}del"
53
+ if len(ref) == 1 and len(alt) == 1:
54
+ return f"c.{pos}{ref}>{alt}"
55
+ if len(ref) == 1 and alt.startswith(ref):
56
+ # e.g. ampC_T-14TGT: GT inserted after the T at -14
57
+ return f"c.{pos}_{_shift(pos, 1)}ins{alt[1:]}"
58
+ return f"c.{span}delins{alt}"
59
+
19
60
 
20
61
  class AmrFinderPlusIterator(hAMRonizedResultIterator):
21
62
 
@@ -93,25 +134,29 @@ class AmrFinderPlusIterator(hAMRonizedResultIterator):
93
134
  skipped_truncated += 1
94
135
  continue
95
136
 
96
- # "POINT" indicates mutational resistance
137
+ # "POINT"/"POINT_DISRUPT" indicate mutational resistance
97
138
  # amrfinderplus has no special fields but the mutation itself is
98
139
  # appended to the symbol name so we want to split this
99
140
  result['amino_acid_mutation'] = None
100
141
  result['nucleotide_mutation'] = None
101
142
  result['genetic_variation_type'] = GENE_PRESENCE
102
143
 
103
- if result['Subtype'] == "POINT":
104
- gene_symbol, mutation = result['Element symbol'].rsplit("_", 1)
105
- result['Element symbol'] = gene_symbol
106
- _, ref, pos, alt, _ = re.split(r"(\D+)(\d+)(\D+)", mutation)
107
- # this means it is a protein mutation
108
- if result['Method'] in ["POINTX", "POINTP"]:
109
- result['amino_acid_mutation'] = f"p.{ref}{pos}{alt}"
110
- result['genetic_variation_type'] = AMINO_ACID_VARIANT
111
- elif result['Method'] == "POINTN":
112
- # e.g., 23S_G2032G ampC_C-11C -> c.2032G>G
113
- result['nucleotide_mutation'] = f"c.{pos}{ref}>{alt}"
114
- result['genetic_variation_type'] = NUCLEOTIDE_VARIANT
144
+ if result['Subtype'] in POINT_SUBTYPES:
145
+ match = POINT_MUTATION_RE.match(result['Element symbol'])
146
+ if match is None:
147
+ warnings.warn(f"Could not parse mutation from element symbol "
148
+ f"{result['Element symbol']} in "
149
+ f"{self.metadata['input_file_name']}")
150
+ else:
151
+ result['Element symbol'] = match['gene']
152
+ # this means it is a protein mutation
153
+ if result['Method'] in ["POINTX", "POINTP"]:
154
+ result['amino_acid_mutation'] = _protein_mutation(match)
155
+ result['genetic_variation_type'] = AMINO_ACID_VARIANT
156
+ elif result['Method'] == "POINTN":
157
+ # e.g., 23S_G2032G -> c.2032G>G, ampC_C-11C -> c.-11C>C
158
+ result['nucleotide_mutation'] = _nucleotide_mutation(match)
159
+ result['genetic_variation_type'] = NUCLEOTIDE_VARIANT
115
160
 
116
161
  # Determine the field_map to use depending on the method used
117
162
  # The following seems to cover all bases with a minimum of fuss