Journal Watch · Bioinformatics · 2026

Integrating curation into scientific publishing to train AI models.

The SourceData-NLP dataset and code are available at https://github.com/source-data/soda-data, https://github.com/source-data/soda-model, and https://github.com/source-data/soda_image_segmentation.

Abreu-Vicente J, Sonntag H, Eidens T et al.Bioinformatics3 min readPharma relevance 50/100

DOI: 10.1093/bioinformatics/btaf685PMID: 41455082Study type: in-vitro / cell study

TL;DR

The SourceData-NLP dataset and code are available at https://github.com/source-data/soda-data, https://github.com/source-data/soda-model, and https://github.com/source-data/soda_image_segmentation.

Key findings

  • We evaluate the utility of the dataset for training AI models using named-entity recognition, segmentation of figure captions into their constituent panels, and a novel context-dependent semantic task that assesses whether an entity is a controlled intervention target or a measurement object.
  • We also demonstrate multi-modal applications for segmenting figures into panel images and their corresponding captions.<h4>Availability and implementation</h4>Trained models are available at https://huggingface.co/EMBO.
  • The SourceData-NLP dataset and code are available at https://github.com/source-data/soda-data, https://github.com/source-data/soda-model, and https://github.com/source-data/soda_image_segmentation.

What they did

The resulting dataset, SourceData-NLP, comprises over 620 000 annotated biomedical entities, curated from 18 689 figures in 3223 articles in molecular and cell biology. Annotations include eight classes of bioentities (small molecules, gene products, subcellular components, cell lines, cell types, tissues, organisms, and diseases), plus additional classes that delineate the entities' roles in experimental designs and methodologies.

Full abstract (original)

Background: <h4>Motivation</h4>High-throughput extraction and structured labeling of data from academic articles are crucial for enabling downstream machine learning applications and secondary analyses. Current approaches lack integration with the publishing process and comprehensive annotation of experimental roles and methodologies alongside bioentity recognition.<h4>Results</h4>We embedded multimodal data curation into the academic publishing process to annotate segmented figure panels and captions, combining natural language processing with authors' feedback to increase annotation accuracy.

What they did: The resulting dataset, SourceData-NLP, comprises over 620 000 annotated biomedical entities, curated from 18 689 figures in 3223 articles in molecular and cell biology. Annotations include eight classes of bioentities (small molecules, gene products, subcellular components, cell lines, cell types, tissues, organisms, and diseases), plus additional classes that delineate the entities' roles in experimental designs and methodologies.

Results & conclusions: We evaluate the utility of the dataset for training AI models using named-entity recognition, segmentation of figure captions into their constituent panels, and a novel context-dependent semantic task that assesses whether an entity is a controlled intervention target or a measurement object. We also demonstrate multi-modal applications for segmenting figures into panel images and their corresponding captions.<h4>Availability and implementation</h4>Trained models are available at https://huggingface.co/EMBO. The SourceData-NLP dataset and code are available at https://github.com/source-data/soda-data, https://github.com/source-data/soda-model, and https://github.com/source-data/soda_image_segmentation.

Abstract © the authors / Bioinformatics (2026). Reproduced here for discussion and indexing, with attribution. Read the original at https://doi.org/10.1093/bioinformatics/btaf685 · PubMed 41455082.

Why it matters for pharma & biotech

  • Directly relevant to teams in small-molecule drug discovery, AI in drug discovery / diagnostics — the abstract reports new data in that space, so competitive and pipeline teams should read the full text.
  • As earlier-stage work, its near-term value is target validation and assay design rather than clinical decisions.
  • For searchers: the full abstract is reproduced below, so you can judge relevance before opening the publisher page.

Modalities & themes detected: small-molecule drug discovery · AI in drug discovery / diagnostics

Limitations & what to check in the full text

  • This brief is based on the published abstract only. Effect sizes, subgroup results and safety detail in the full text may qualify the headline findings.
  • The work is preclinical. Results in animal or cell models frequently do not translate directly to human efficacy or safety.
  • The abstract reports its conclusions qualitatively, without headline effect sizes — read the full text before citing a magnitude.

Frequently asked questions

What did this Bioinformatics study find?

The SourceData-NLP dataset and code are available at https://github.com/source-data/soda-data, https://github.com/source-data/soda-model, and https://github.com/source-data/soda_image_segmentation.

What type of study is this?

It is a in-vitro / cell study, published in Bioinformatics in 2026. The resulting dataset, SourceData-NLP, comprises over 620 000 annotated biomedical entities, curated from 18 689 figures in 3223 articles in molecular and cell biology.

Where can I read the original paper?

The original is at the publisher via DOI 10.1093/bioinformatics/btaf685. PubMed ID: 41455082. BioStudioAI reproduces the abstract for discussion and links to the source.

How to cite the original: Abreu-Vicente J, Sonntag H, Eidens T et al.. Integrating curation into scientific publishing to train AI models.. Bioinformatics (2026). DOI: 10.1093/bioinformatics/btaf685.
This brief: BioStudioAI Journal Watch, 29 October 2026. https://biostudioai.com/blog/papers/integrating-curation-into-scientific-publishing-to-train-ai-models-taf685/

Reading papers is step one. Analysing the data is the work.

BioStudioAI turns raw lab files into QC'd results, statistics, figures, reports and slides — hosted on your servers, every number computed, nothing typed.

Open BioStudioAI More guides