Towards an ontology-based recommender system for relevant bioinformatics workflows
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Abstract Background With the large and diverse type of biological data, bioinformatic solutions are being more complex and computationally intensive. New specialized data skills need to be acquired by researchers in order to follow this development. Workflow Management Systems rise as an efficient way to automate tasks through abstract models in order to assist users during their problem solving tasks. However, current solutions could have several problems in reusing the developed models for given tasks. The large amount of heterogenous data and the lack of knowledge in using bioinformatics tools could mislead the users during their analyses. To tackle this issue, we propose an ontology-based workflow-mining framework generating semantic models of bioinformatic best practices in order to assist scientists. To this end, concrete workflows are extracted from scientific articles and then mined using a rich domain ontology. Results In this study, we explore the specific topics of phylogenetic analyses. We annotated more than 300 recent articles using different ontological concepts and relations. Relative supports (frequencies) of discovered workflow components in texts show interesting results of relevant resources currently used in the different phylogenetic analysis steps. Mining concrete workflows from texts lead us to discover abstract but relevant patterns of the best combinations of tools, parameters and input data for specific phylogenetic problems. Conclusions Extracted patterns would make workflows more intuitive and easy to be reused in similar situations. This could provide a stepping-stone into the identification of best practices and pave the road to a recommender system.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.012 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.000 |
| Open science | 0.004 | 0.002 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it