SIDER - Side Effect Resource

SkillProductivity

Category: Drug-centric | Type: DB | Subcategory: Adverse Drug Reaction (ADR) Link: http://sideeffects.embl.de/ | Paper: https://doi.org/10.1093/nar/gkv1075

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the SIDER - Side Effect Resource skill

About this skill

🦀 Agentic RAG for drug intelligence · 57 skills · 15 task categories · DTI · ADR · DDI · PGx · Repurposing · Powered by LangGraph

What this skill tells your AI

The instructions your AI receives, as published by qsong-github/drugclaw in skills/adr/sider/SKILL.md and read by ahel’s review.

Category: Drug-centric | Type: DB | Subcategory: Adverse Drug Reaction (ADR) Link: http://sideeffects.embl.de/ | Paper: https://doi.org/10.1093/nar/gkv1075

SIDER contains marketed medicines and their recorded adverse drug reactions, mined from public documents and package inserts.

Entity Type Auto-Detection

Input PatternDetected AsExample
CIDxxxxxxxxxSTITCH IDCID100000085
Cxxxxxxx (8-char, digits after C)UMLS CUIC0011849
X##XX## (7-char ATC pattern)ATC codeN05BA01
anything elsefree textaspirin, headache

Free text first attempts exact/substring match on drug names; if no drug is matched, it searches side-effect and indication names.

API

FunctionInputReturns
search(entity)single entity stringdict with drug_name, stitch_ids, atc_codes, side_effects, indications, freq
search_batch(entities)list of entity stringsdict[str, dict]
summarize(result)one result dictcompact text string
to_json(result)one result dictsame dict (pipeline passthrough)

Usage

from 14_SIDER import search, search_batch, summarize

# Single query — drug name
result = search("aspirin")
print(summarize(result))

# Single query — STITCH ID
result = search("CID100000085")
print(summarize(result))

# Single query — side-effect text
result = search("headache")
print(summarize(result))

# Batch query — mixed entity types
results = search_batch(["metformin", "ibuprofen", "C0011849"])
for entity, r in results.items():
    print(summarize(r))

Data Files

All pre-downloaded and extracted under: resources_metadata/adr/SIDER

FileContent
drug_names.tsvSTITCH flat ID → drug name
drug_atc.tsvSTITCH flat ID → ATC code
meddra_all_se.tsvAll side effects (STITCH, UMLS, MedDRA type, SE name)
meddra_all_indications.tsvAll indications (STITCH, UMLS, method, MedDRA name)
meddra_freq.tsvSide effects with frequency info

The loader auto-detects .tsv / .tsv.gz variants. Results are capped at 50 records per category for readability.

Signals

GitHub stars
116
Forks
3
Last commit
Aug 2026
Advanced
Item type
skill
Key
sider
Source
github.com/qsong-github/drugclaw