Automated Extraction of Protein Mutation Impacts from the Biomedical Literature
Bibliographic record
Abstract
Mutations as sources of evolution have long been the focus of attention in the \nbiomedical literature. Accessing the mutational information and their impacts \non protein properties facilitates research in various domains, such as \nenzymology and pharmacology. However, manually reading through the rich and fast growing repository \nof biomedical literature is expensive and time-consuming. A number of manually curated \ndatabases, such as BRENDA (http://www.brenda-enzymes.org), try to index and provide this \ninformation; yet the provided data seems to be incomplete. Thus, there is a \ngrowing need for automated approaches to extract this information. \nIn this work, we present a system to automatically extract and summarize impact \ninformation from protein mutations. \nOur system extraction module is split into subtasks: organism analysis, \nmutation detection, protein property extraction and impact \nanalysis. Organisms, as sources of proteins, are required to be extracted to \nhelp disambiguation of genes and proteins. Thus, our system extracts and \ngrounds organisms to NCBI. We detect mutation series to correctly ground our detected \nimpacts. Our system also extracts the affected protein properties as well as the magnitude of the \neffects. \nThe output of our system is populated to an OWL-DL ontology, which can then be queried to provide structured information. The performance \n of the system is evaluated on both external and internal corpora and \n databases. The results show the reliability of the approaches. Our Organism \n extraction system achieves a precision and recall of 95% \nand 94% and a grounding accuracy of 97.5% on the OT corpus. On the manually \nannotated corpus of Linneaus-100, the results show a precision and recall of \n99% and 97% and grounding with an accuracy of 97.4%. \nIn the impact detection task, our system achieves a precision and recall of \n70.4%-71.8% and 71.2%-71.3% on a manually annotated documents. Our system grounds the detected \nimpacts with an accuracy of 70.1%-71.7% on the manually annotated documents \nand a precision and recall of 57%-57.5% and 82.5%-84.2% against the BRENDA data.
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 | 0.002 | 0.010 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.034 | 0.010 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.007 |
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".