Putting benchmarks in their rightful place: The heart of computational biology
Bibliographic record
Abstract
Research in computational biology has given rise to a vast number of methods developed to solve scientific problems.For areas in which many approaches exist, researchers have a hard time deciding which tool to select to address a scientific challenge, as essentially all publications introducing a new method will claim better performance than all others.Not all of these claims can be correct.Equally, for this same reason, developers struggle to demonstrate convincingly that they created a new and superior algorithm or implementation.Moreover, the developer community often has difficulty discerning which new approaches constitute true scientific advances for the field.The obvious answer to this conundrum is to develop benchmarks-meaning standard points of reference that facilitate evaluating the performance of different tools-allowing both users and developers to compare multiple tools in an unbiased fashion.Broadly speaking, benchmarks consist of input data that methods are meant to operate upon, expected output data against which tool output can be compared, a specification of metrics used to assess performance, and performance values of sets of tools that have been run through the benchmark.Developing good and comprehensive benchmarks, in which the performance metrics of each tool reflect its real-world utility, requires a significant effort.For highly competitive and established fields, such as protein structure predictions, community experiments evaluating the methods have been held periodically to provide blinded assessments of prediction performance.These blinded assessments are perhaps the gold standard on how benchmarks should be run.However, in most areas of computational biology, no such regular blinded contests are available.Instead, many tool developers end up generating their own benchmarks, which they publish alongside a newly developed tool to show its improved performance.The downside of this approach is that, if a new approach is developed in parallel to assembly of the benchmark on which it is evaluated, there is a strong selection bias encouraging the authors to report tool development approaches performing well against the benchmark compared to previous tools.This reporting bias makes most benchmarks that accompany newly developed tools questionable.Even if the authors are aware of this problem and take conscious steps to separate
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 | 0.029 | 0.128 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.003 | 0.008 |
| Scholarly communication | 0.011 | 0.011 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.011 | 0.035 |
| Insufficient payload (model declined to judge) | 0.004 | 0.004 |
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".