literature-review

Define a research question using the PICO framework, then search at least three databases such as PubMed, bioRxiv, and Semantic Scholar. Screen results through title, abstract, and full-text stages, synthesize findings thematically, and verify all citations before generating the final document.

Run the verify_citations.py script on your markdown file to extract every DOI, check each one against the CrossRef API, and retrieve correct metadata. It produces a verification report listing failed DOIs and generates properly formatted citations in APA or Nature style.

Search a minimum of three complementary sources: PubMed for peer-reviewed biomedical literature, bioRxiv or medRxiv for recent preprints, and Semantic Scholar or Google Scholar for cross-disciplinary coverage. Specialized databases like ChEMBL, UniProt, or COSMIC suit domain-specific topics.

Yes, PDF generation requires pandoc and a LaTeX engine such as xelatex installed on your system. You can check both dependencies by running generate_pdf.py with the --check-deps flag before converting your markdown review.

Thematic organization synthesizes findings across multiple studies, revealing consensus, contradictions, and research gaps. Study-by-study summaries merely list results without critical analysis, which fails academic review standards and PRISMA guidelines.