Review papers with raw data transparency

Paper review in one paragraph

The bibliometric study maps 828 Scopus-indexed original articles (2008–2024) on miRNAs in colorectal cancer and identifies growth phases, leading countries and institutions, frequently studied miRNAs (miR-21, miR-34a, miR-195-5p), dominant themes (proliferation, EMT, chemoresistance) and emerging frontiers (ferroptosis, ceRNA, exosome biology); data and pipelines are publicly available on GitHub and the authors note primary limitations including single database (Scopus) and English-only inclusion Bibliometric study summary

Detailed critical review and critique

Paper identification

Title: Exploring miRNA Research in Colorectal Cancer: Insights from a Bibliometric Analysis — Pharmaceutics 2025 (DOI 10.3390/pharmaceutics17081084) — 828 included articles; GitHub pipeline available.

Paper metadata and core methods

What the paper did well

• Transparent dataset and code availability: authors published their raw Scopus export and Python/R pipelines on GitHub, enabling reanalysis and reproducibility Code and data availability on GitHub
• Methodological clarity: authors documented search syntax, screening rules (titles/keywords restricted), and used validated tools (bibliometrix, VOSviewer, miRBase) which increases methodological transparency Search strategy and toolchain
• Useful domain synthesis: identification of core miRNAs (miR-21, miR-34a, miR-195-5p), dominant biological themes (proliferation, EMT, chemoresistance), and emerging topics (ferroptosis, ceRNA, exosomes) provides practical signposts for targeted meta-analyses and experimental prioritization Thematic and topic identification

Main limitations, biases, and blindspots

1. Single-source bias: exclusive reliance on Scopus is explicitly acknowledged by the authors and can skew country, journal, and citation patterns relative to combined datasets (Scopus vs Web of Science vs PubMed). This matters because different databases index different journals and regional literature; for example, some local-language or regional journals may be underrepresented in Scopus leading to geographic bias Limitations of single-database retrieval
2. Language and document-type filters: restricting to English and original research (excluding reviews/meta-analyses) removes synthesis-level publications and non-English studies that could change thematic weighting or identify historically important works in other languages Language and article-type restrictions
3. Search strategy sensitivity/specificity tradeoff: authors limited retrieval to TITLE and author KEYWORDS (mir-* patterns). This reduces off-topic noise but risks missing articles that discuss miRNAs only in abstracts or full text; some valid miRNA-CRC studies could be lost, biasing counts especially for miRNAs with nonstandard naming or earlier nomenclature before miRBase harmonization Search field restrictions
4. miRNA extraction and nomenclature change risk: although validation against miRBase releases 20–22 mitigates misidentification, miRNA nomenclature has evolved; older studies may use obsolete names or precursor IDs, potentially undercounting earlier signals; the paper reports 484 miRNA IDs extracted but does not show sensitivity analyses for alternate miRBase releases or fuzzy matching strategies miRNA extraction and validation procedure
5. Counting and attribution effects for country metrics: the paper reports China as top contributor (high SCPs) but also low MCP proportion for Chinese corresponding authors; simple SCP counts can obscure multi-institution, multi-author contributions and the impact-quality tradeoff (many publications may have low citation impact); the authors present both global and local citation metrics which helps, but further normalization by per-article citation rates or journal impact could refine conclusions Country contributions and MCP/SCP analysis

Interpretation validity and strength of claims

The authors' core descriptive claims are well supported by the dataset they curated: publication counts, identified miRNAs (miR-21, miR-34a, miR-195-5p), and dominant keywords follow directly from extraction and frequency analyses performed with bibliometrix/VOSviewer and the custom Python script miRNA frequency results. However, causal or mechanistic claims (for instance implying translational readiness of certain miRNAs) would be outside bibliometrics and require separate experimental meta-analyses and functional validation; the authors correctly limit such claims and propose follow-up targeted meta-analyses and functional validation steps Future directions and cautious translation.

Concrete suggestions to improve the study and follow-up work

• Integrate additional bibliographic sources (Web of Science, PubMed/MEDLINE) and perform deduplication to reduce single-source bias and capture regional/non-English literature.
• Include reviews/meta-analyses as a separate stratum to reveal synthesis-level influence and how primary research fed into clinical translational thinking.
• Run sensitivity analyses for miRNA name variants using historical miRBase releases and fuzzy matching to quantify potential undercounts from nomenclature drift.
• Report normalized impact measures (citations per paper, field-weighted citation impact) to complement volume metrics and highlight high-impact but smaller-output contributors.
• Make available metadata tables (per-article miRNA mentions, citation counts, country affiliations) in machine-readable format to facilitate downstream meta-analyses and targeted systematic reviews.

Practical value to researchers

This study is a practical roadmap for: (1) selecting candidate miRNAs for targeted meta-analysis (start with miR-21, miR-34a, miR-195-5p); (2) prioritizing experimental validation for chemotherapy resistance mechanisms (oxaliplatin and 5-FU-related miRNA interactions); (3) exploring emerging mechanistic areas (ferroptosis, ceRNA, exosome-mediated miRNA transport) that are less populated but rising.

Confidence and falsifiability

Confidence in the descriptive bibliometric results is moderate to high given available data and transparent code; key claims are falsifiable by reproducing the analysis with alternative databases and search fields (if top miRNAs, themes, or country rankings differ markedly, conclusions would need revision) Falsifiability statement

Interactive next steps

If you want, I can:

• Reproduce the authors analyses (temporal trends, top miRNAs, country maps) from their GitHub export and produce reproducible figures or alternative analyses integrating PubMed/WoS (requires running the bioinformatics agent below).
• Generate a prioritized list of miRNAs for targeted meta-analysis with suggested inclusion criteria and search strings.
• Run sensitivity testing on miRNA extraction using earlier miRBase releases and fuzzy matching to quantify nomenclature effects.

Note: Data and code are available at the authors GitHub which facilitates full reproducibility: https://github.com/MichelangeloAloisio/Bibliometric_Analysis_miRNA_CRC/ GitHub data link