scienceclaw-ie

Files (1)

• SKILL.md

SKILL.md (first lines)

# ScienceCLAW Information Extraction

Extract structured information from scientific texts including named entities, relations, data tables, methods, results, and experimental parameters.

## When to Use

Use this skill when the user:

- Needs to parse a scientific paper for key data points (measurements, parameters, outcomes)
- Wants to extract experimental methods, conditions, or protocols from a paper
- Needs named entity recognition (NER) on scientific documents (chemicals, genes, diseases, etc.)
- Wants to build a knowledge graph or structured database from literature
- Needs to extract relations between scientific entities (drug-target, gene-disease, cause-effect)
- Wants structured tables from unstructured scientific text
- Needs to compare methods or results across multiple papers in a structured format
- Wants to extract metadata from papers (authors, affiliations, funding, conflicts of interest)

## When NOT to Use

Do not use this skill when:

- The user wants a narrative summary of a paper (use scienceclaw-summarization)
- The user needs to find and retrieve papers (use scienceclaw-retrieval)
- The user wants to answer a scientific question (use scienceclaw-qa)
- The user needs multi-step reasoning or proof construction (use scienceclaw-reasoning)
- The user needs code-based data analysis (use code-execution)

## Entity Types by Discipline

### Biomedical Sciences

| Entity Type | Examples | Common Formats |
|---|---|---|
| Gene/Protein | TP53, BRCA1, insulin | Uppercase symbols, UniProt IDs |
| Disease | Type 2 diabetes, glioblastoma | MeSH terms, ICD codes |
| Drug/Chemical | Metformin, aspirin, NaCl | IUPAC names, brand names, CAS numbers |
| Organism | Homo sapiens, E. coli | Binomial nomenclature, NCBI Taxonomy ID |
| Cell Line | HeLa, HEK293T | ATCC identifiers |
| Anatomical Structure | hippocampus, mitochondria | UBERON ontology terms |
| Biological Process | apoptosis, glycolysis | GO terms |
| Dosage | 10 mg/kg, 500 nM | Value + unit patterns |
| Biomarker | HbA1c, PSA, CRP | Abbreviations with reference ranges |

### Chemistry and Materials Science

| Entity Type | Examples | Common Formats |
|---|---|---|
| Chemical Compound | benzene, polyethylene glycol | IUPAC, SMILES, InChI |
| Material | graphene, Ti-6Al-4V, PDMS | Common names, composition formulas |
| Property | melting point, band gap, tensile strength | Property name + value + unit |
| Synthesis Method | sol-gel, CVD, electrospinning | Method name + parameters |
| Characterization Technique | XRD, SEM, NMR, FTIR | Acronyms with parameters |
| Crystal Structure | FCC, BCC, hexagonal | Space group, lattice parameters |
| Catalyst | Pd/C, TiO2, zeolite | Composition + support |

### Physics and Astronomy

| Entity Type | Examples | Common Formats |
|---|---|---|
| Physical Quantity | 3.14 eV, 300 K, 1.5 T | Value + SI/derived unit |
| Particle | electron, muon, Higgs boson | Standard Model names |
| Astronomical Object | NGC 1277, Proxima Centauri | Catalog designations |
| Physical Constant | c, h, G, k_B | Symbol + numerical value + unit |
| Equation/Law | Schrodinger equation, F = ma | Named equations, formulas |
| Experimental Apparatus | LHC, LIGO, HST | Acronyms and facility names |
| Measurement Technique | spectroscopy, interferometry | Technique name + configuration |

### Social Sciences

| Entity Type | Examples | Common Formats |
|---|---|---|
| Study Design | RCT, cohort study, survey | Methodology descriptors |
| Statistical Measure | p = 0.03, r = 0.45, d = 0.8 | Statistic symbol + value |
| Sample | N = 1,200 adults aged 18-65 | Size + demographic descriptors |
| Instrument/Scale | Likert scale, BDI-II, MMSE | Named instruments with citations |
| Effect | income inequality, voter turnout | Outcome variable descriptions |
| Theory/Framework | rational choice, social learning | Named theoretical frameworks |
| Policy/Intervention | minimum wage increase, CBT | Intervention descriptions |
| Geographic Scope | United States, OECD countries | Country/region names |
| Time Period | 2010-2020, post-WWII | Date ranges, era references |

### Computer Science

| Entity Type | Examples | Common Formats |
|---|---|---|
| Algorithm | Adam optimizer, BERT, ResNet-50 | Named algorithms with versions |
| Dataset | ImageNet, MNIST, GLUE | Named benchmarks |
| Metric | F1 = 0.92, accuracy = 95.3%, BLEU = 32.1 | Metric name + value |
| Architecture | transformer, CNN, GAN | Architecture type names |
| Hyperparameter | learning rate = 1e-4, batch size = 32 | Parameter name + value |
| Framework/Library | PyTorch, TensorFlow, scikit-learn | Software names with versions |
| Hardware | A100 GPU, TPU v4 | Hardware specifications |

## Relation Schemas

### Core Relation Types

| Relation | Description | Example |
|---|---|---|
| TREATS | Drug/intervention treats disease | (Metformin, TREATS, Type 2 diabetes) |
| CAUSES | Agent causes effect | (Smoking, CAUSES, Lung cancer) |
| INHIBITS | Entity inhibits another | (Aspirin, INHIBITS, COX-2) |
| ACTIVATES | Entity activates another | (EGF, ACTIVATES, EGFR) |
| PART_OF | Component is part of whole | (Hippocampus, PART_OF, Limbic system) |
| MEASURED_BY | Property measured by technique | (Crystal structure, MEASURED_BY, XRD) |
| SYNTHESIZED_VIA | Material made by method | (Graphene, SYNTHESIZED_VIA, CVD) |
| ASSOCIATED_WITH | Statistical association | (Gene X, ASSOCIATED_WITH, Disease Y) |
| OUTPERFORMS | Method A outperforms B | (BERT, OUTPERFORMS, LSTM on GLUE) |
| DERIVED_FROM | Result derived from data/method | (Estimate, DERIVED_FROM, Meta-analysis) |
| USED_IN | Tool/method used in study | (fMRI, USED_IN, Smith et al. 2023) |
| CONTRADICTS | Finding contradicts another | (Study A, CONTRADICTS, Study B) |
| SUPPORTS | Finding supports another | (Experiment, SUPPORTS, Hypothesis) |

### Relation Attributes

Each extracted relation should include:

- **Source**: The text span or sentence from which the relation was extracted
- **Confidence**: High / Medium / Low based on explicitness of the statement

From beita6969/ScienceClaw, MIT.

More from beita6969/ScienceClaw