Literature Agent v1.0.0
Overview
Dual-source literature search combining PubMed (biomedical focus) and Semantic Scholar (broader CS/ML/AI coverage). Deduplicates across sources, enriches with citation metrics and TLDR summaries.
Key capabilities:
- - PubMed search with MeSH terms, abstracts, publication types
- Semantic Scholar search with citation counts, influential citations, TLDR
- Paper lookup by DOI or PMID
- Citation tracking (who cited this paper?)
- Related paper discovery (what did this paper reference?)
- Automatic query construction from compound/target/disease inputs
- Cross-source deduplication and enrichment
Quick Start
CODEBLOCK0
Scripts
scripts/pubmed_search.py
PubMed via NCBI E-utilities (public, no key required, rate limit: 3 req/sec).
CODEBLOCK1
Returns: PMID, title, authors, journal, year, DOI, abstract, MeSH terms, keywords, publication types.
scripts/semantic_scholar.py
Semantic Scholar API (public, no key required, rate limit: 100 req/5 min).
CODEBLOCK2
Returns: title, authors, year, abstract, TLDR, citation count, influential citations, DOI, ArXiv ID, open-access PDF URL.
scripts/chain_entry.py
Standard PharmaClaw chain interface. Searches both PubMed and Semantic Scholar, deduplicates, and sorts by citation impact.
Input keys: query, compound/name, target, disease, mechanism, reaction, topic, doi, pmid, max_results, years, INLINECODE15
Automatic query building: {"compound": "aspirin", "disease": "colorectal cancer"} → searches "aspirin colorectal cancer"
Chaining
| From | Input | To |
|---|
| Chemistry Query | Compound name/SMILES | Literature → find published studies |
| Catalyst Design |
Reaction type |
Literature → find catalyst optimization papers |
|
Literature | Key findings | Pharmacology → validate claims |
|
Literature | Synthesis references | Chemistry Query → retrosynthesis |
|
Literature | Patent mentions | IP Expansion → FTO analysis |
文献智能体 v1.0.0
概述
双源文献检索,结合PubMed(生物医学重点)和Semantic Scholar(更广泛的计算机科学/机器学习/人工智能覆盖范围)。跨源去重,并丰富引用指标和TLDR摘要。
关键能力:
- - PubMed检索,支持MeSH术语、摘要、出版物类型
- Semantic Scholar检索,支持引用计数、高影响力引用、TLDR
- 通过DOI或PMID查找论文
- 引用追踪(谁引用了这篇论文?)
- 相关论文发现(这篇论文引用了什么?)
- 从化合物/靶点/疾病输入自动构建查询
- 跨源去重和丰富
快速开始
bash
按主题搜索
python scripts/pubmed_search.py --query KRAS G12C抑制剂 --max-results 5
搜索Semantic Scholar(包括机器学习/人工智能论文)
python scripts/semantic_scholar.py --query 图神经网络药物发现
完整链:化合物 + 疾病背景
python scripts/chain_entry.py --input-json {compound: 索托拉西布, disease: 肺癌}
查找特定论文并找到引用它的文献
python scripts/semantic_scholar.py --paper-id DOI:10.1038/s41586-021-03819-2 --citations
仅最近论文(最近3年)
python scripts/pubmed_search.py --query 有机金属催化剂药物合成 --years 3
脚本
scripts/pubmed_search.py
通过NCBI E-utilities进行PubMed检索(公开,无需密钥,速率限制:3次请求/秒)。
--query <文本> 必需。搜索查询
--max-results 1-50(默认:10)
--sort <类型> 相关性 | 日期(默认:相关性)
--years 限制最近N年
返回:PMID、标题、作者、期刊、年份、DOI、摘要、MeSH术语、关键词、出版物类型。
scripts/semantic_scholar.py
Semantic Scholar API(公开,无需密钥,速率限制:100次请求/5分钟)。
--query <文本> 搜索查询
--paper-id 论文ID(DOI:xxx、PMID:xxx、ArXiv:xxx)
--related 获取论文的参考文献(需要--paper-id)
--citations 获取引用论文的文献(需要--paper-id)
--max-results 1-50(默认:10)
--year-range <范围> 例如2020-2026或2023-
返回:标题、作者、年份、摘要、TLDR、引用计数、高影响力引用、DOI、ArXiv ID、开放获取PDF URL。
scripts/chain_entry.py
标准PharmaClaw链接口。同时搜索PubMed和Semantic Scholar,去重,并按引用影响力排序。
输入键:query、compound/name、target、disease、mechanism、reaction、topic、doi、pmid、max_results、years、context
自动查询构建:{compound: 阿司匹林, disease: 结直肠癌} → 搜索阿司匹林结直肠癌
链式连接
| 来源 | 输入 | 输出 |
|---|
| 化学查询 | 化合物名称/SMILES | 文献 → 查找已发表研究 |
| 催化剂设计 |
反应类型 |
文献 → 查找催化剂优化论文 |
|
文献 | 关键发现 | 药理学 → 验证声明 |
|
文献 | 合成参考文献 | 化学查询 → 逆合成分析 |
|
文献 | 专利提及 | 知识产权扩展 → 自由实施分析 |