ChatGPT usage in the Reactome curation process
Bibliographic record
Abstract
Appreciating the rapid advancement and ubiquity of generative AI, particularly ChatGPT, a chatbot using large language models like GPT, we endeavour to explore the potential application of ChatGPT in the data collection and annotation stages within the Reactome curation process. This exploration aimed to create an automated or semi-automated framework to mitigate the extensive manual effort traditionally required for gathering and annotating information pertaining to biological pathways, adopting a Reactome "reaction-centric" approach. In this pilot study, we used ChatGPT/GPT4 to address gaps in the pathway annotation and enrichment in parallel with the conventional manual curation process. This approach facilitated a comparative analysis, where we assessed the outputs generated by ChatGPT against manually extracted information. The primary objective of this comparison was to ascertain the efficiency of integrating ChatGPT or other large language models into the Reactome curation workflow and helping plan our annotation pipeline, ultimately improving our protein-to-pathway association in a reliable and automated or semi-automated way. In the process, we identified some promising capabilities and inherent challenges associated with the utilisation of ChatGPT/GPT4 in general and also specifically in the context of Reactome curation processes. We describe approaches and tools for refining the output given by ChatGPT/GPT4 that aid in generating more accurate and detailed output.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.033 | 0.081 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.004 | 0.008 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.014 | 0.012 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".