Accurate prediction of disorder in protein chains with a comprehensive and empirically designed consensus
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
Availability of computational methods that predict disorder from protein sequences fuels rapid advancements in the protein disorder field. The most accurate predictions are usually obtained with consensus-based approaches. However, their design is performed in an ad hoc manner. We perform first-of-its-kind rational design where we empirically search for an optimal mixture of base methods, selected out of a comprehensive set of 20 modern predictors, and we explore several novel ways to build the consensus. Our method for the prediction of disorder based on Consensus of Predictors (disCoP) combines seven base methods, utilizes custom-designed set of selected 11 features that aggregate base predictions over a sequence window and uses binomial deviance loss-based regression to implement the consensus. Empirical tests performed on an independent benchmark set (with low-sequence similarity compared with proteins used to design disCoP), shows that disCoP provides statistically significant improvements with at least moderate magnitude of differences. disCoP outperforms 28 predictors, including other state-of-the-art consensuses, and achieves Area Under the ROC Curve of .85 and Matthews Correlation Coefficient of .5 compared with .83 and .48 of the best considered approach, respectively. Our consensus provides high rate of correct disorder predictions, especially when low rate of incorrect disorder predictions is desired. We are first to comprehensively assess predictions in the context of several functional types of disorder and we demonstrate that disCoP generates accurate predictions of disorder located at the post-translational modification sites (in particular phosphorylation sites) and in autoregulatory and flexible linker regions. disCoP is available at http://biomine.ece.ualberta.ca/disCoP/.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 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