phee

Name: phee
Author: QSong-github

132stars4forks

How to install phee?

qsong-github/drugclaw/phee

$npx -y skills add QSong-github/DrugClaw --skill phee

Installs into the current project.

Summary

Query the PHEE pharmacovigilance event extraction dataset. Use whenever the user asks about annotated adverse drug events, pharmacovigilance case reports, drug–effect associations from medical literature, or wants to find PHEE examples mentioning a drug name, adverse effect, or c

Files · 1

View on GitHub

SKILL.md

1	# PHEE Query Skill
2
3	Search PHEE annotated pharmacovigilance events by drug name, adverse effect,
4	or free text. Substring matching is case-insensitive across all event fields.
5
6	\| Input Pattern \| Match Logic \|
7	\|---\|---\|
8	\| drug name (e.g. `phenytoin`) \| substring on `drug`, `treatment`, `combination_drug`, and `text` \|
9	\| adverse effect (e.g. `hepatotoxicity`) \| substring on `effect`, `treatment_disorder`, and `text` \|
10	\| any other text \| substring on all event fields and full sentence \|
11
12	## API
13
14	\| Function \| Input \| Returns \|
15	\|---\|---\|---\|
16	\| `load_phee(data_dir)` \| directory path \| list[dict] (cached) \|
17	\| `search(records, entity)` \| single entity string \| list[dict] \|
18	\| `search_batch(records, entities)` \| list of entity strings \| dict[str, list[dict]] \|
19	\| `summarize(hits, entity)` \| hit list + label \| compact text (sentence count, ADE/PTE counts, top drugs & effects) \|
20	\| `to_json(hits)` \| hit list \| list[dict] \|
21
22	## Record Schema
23
24	Each record returned by `search`:
25
26	```
27	{
28	"id": "8908396_3",
29	"text": "A 52-year-old Black woman on phenytoin therapy ...",
30	"split": "train",
31	"events": [
32	{
33	"event_type": "Adverse_event \| Potential_therapeutic_event",
34	"trigger": ["developed"],
35	"subject": ["A 52-year-old Black woman ..."],
36	"subject_age": ["52-year-old"],
37	"subject_gender": ["woman"],
38	"subject_race": ["Black"],
39	"subject_population": [],
40	"subject_disorder": ["post-traumatic epilepsy"],
41	"treatment": ["phenytoin therapy ..."],
42	"drug": ["phenytoin"],
43	"dosage": [],
44	"freq": [],
45	"route": [],
46	"duration": [],
47	"treatment_disorder": ["post-traumatic epilepsy"],
48	"combination_drug": [],
49	"effect": ["transient hemiparesis"]
50	}
51	]
52	}
53	```
54
55	## Usage
56
57	See `if __name__ == "__main__"` block in `38_PHEE.py` for runnable examples
58	covering: single drug search, single effect search, batch search, and JSON
59	output.
60
61	## Data
62
63	- Source: PHEE dataset (Zenodo record 7689970), EMNLP 2022
64	- Files: `train.json` (~2,900 sentences), `dev.json` (~960), `test.json` (~960)
65	- Path: `DATA_DIR` variable in `38_PHEE.py`
66	(`resources_metadata/drug_nlp/PHEE/data/json`)
67	- Event types: `Adverse_event`, `Potential_therapeutic_event`
68	- Main arguments: Subject, Treatment, Effect
69	- Sub-arguments: Subject (age, gender, race, population, disorder); Treatment (drug, dosage, freq, route, duration, disorder, combination.drug)

People who liked this also viewed

6 picks

Type

phee

132stars4forks

View on GitHub

How to install phee?

qsong-github/drugclaw/phee

$npx -y skills add QSong-github/DrugClaw --skill phee

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# PHEE Query Skill
2
3	Search PHEE annotated pharmacovigilance events by drug name, adverse effect,
4	or free text. Substring matching is case-insensitive across all event fields.
5
6	\| Input Pattern \| Match Logic \|
7	\|---\|---\|
8	\| drug name (e.g. `phenytoin`) \| substring on `drug`, `treatment`, `combination_drug`, and `text` \|
9	\| adverse effect (e.g. `hepatotoxicity`) \| substring on `effect`, `treatment_disorder`, and `text` \|
10	\| any other text \| substring on all event fields and full sentence \|
11
12	## API
13
14	\| Function \| Input \| Returns \|
15	\|---\|---\|---\|
16	\| `load_phee(data_dir)` \| directory path \| list[dict] (cached) \|
17	\| `search(records, entity)` \| single entity string \| list[dict] \|
18	\| `search_batch(records, entities)` \| list of entity strings \| dict[str, list[dict]] \|
19	\| `summarize(hits, entity)` \| hit list + label \| compact text (sentence count, ADE/PTE counts, top drugs & effects) \|
20	\| `to_json(hits)` \| hit list \| list[dict] \|
21
22	## Record Schema
23
24	Each record returned by `search`:
25
26	```
27	{
28	"id": "8908396_3",
29	"text": "A 52-year-old Black woman on phenytoin therapy ...",
30	"split": "train",
31	"events": [
32	{
33	"event_type": "Adverse_event \| Potential_therapeutic_event",
34	"trigger": ["developed"],
35	"subject": ["A 52-year-old Black woman ..."],
36	"subject_age": ["52-year-old"],
37	"subject_gender": ["woman"],
38	"subject_race": ["Black"],
39	"subject_population": [],
40	"subject_disorder": ["post-traumatic epilepsy"],
41	"treatment": ["phenytoin therapy ..."],
42	"drug": ["phenytoin"],
43	"dosage": [],
44	"freq": [],
45	"route": [],
46	"duration": [],
47	"treatment_disorder": ["post-traumatic epilepsy"],
48	"combination_drug": [],
49	"effect": ["transient hemiparesis"]
50	}
51	]
52	}
53	```
54
55	## Usage
56
57	See `if __name__ == "__main__"` block in `38_PHEE.py` for runnable examples
58	covering: single drug search, single effect search, batch search, and JSON
59	output.
60
61	## Data
62
63	- Source: PHEE dataset (Zenodo record 7689970), EMNLP 2022
64	- Files: `train.json` (~2,900 sentences), `dev.json` (~960), `test.json` (~960)
65	- Path: `DATA_DIR` variable in `38_PHEE.py`
66	(`resources_metadata/drug_nlp/PHEE/data/json`)
67	- Event types: `Adverse_event`, `Potential_therapeutic_event`
68	- Main arguments: Subject, Treatment, Effect
69	- Sub-arguments: Subject (age, gender, race, population, disorder); Treatment (drug, dosage, freq, route, duration, disorder, combination.drug)

People who liked this also viewed

6 picks

Type