Artificiell Intelligens-baserad förutsägelse av proteinkomplex formade via flyktiga interaktioner
Bibliographic record
Abstract
Transient protein-protein interactions (tPPIs) are interactions between multiple proteins which swiftly switch between an associated and a dissociated state. Such interactions play a crucial role in many dynamical processes in the cell, such as in the regulatory and signaling pathways. Because of their adaptable nature, tPPIs are attractive targets for the development of novel biomaterials. tPPIs have previously been overlooked, due to their low cellular concentrations as well as being hard to capture both experimentally and with computational methods. As a result, there exists a knowledge gap of the structures and functions of tPPIs. This thesis provides a comprehensive analysis of predicting tPPIs with the front-runner of the AI-driven structure predicting algorithms, AlphaFold (AF). Prior to this work, no publications have addressed AI-driven structure predictions of tPPIs. The AF generated tPPI-models are compared against experimental data when available, and against the output of the protein docking algorithm ClusPro. The AF performance is further validated through molecular dynamics simulations. AF showed to be much less confident about predicting tPPIs than stable protein complexes in general. High sampling should therefore be employed to acquire a wider range of models. The AF generated models did not significantly differ from those predicted by ClusPro, and were also fairly close to the experimentally determined interaction site. Interaction site families were readily identifiable from the sample by isolating and comparing the residues predicted to be in contact. Simulating models from each family with MD, showed that AF is able to predict a range of stable interaction sites. The interaction site features varied among the studied proteins, but the most commonly predicted interaction sites included beta-strand residues and disordered loops for all studied proteins. The detailing of the benefits and limitations of predicting tPPIs with AF presented in this work might aid in the engineering of new tPPI-based biomaterials.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.014 | 0.011 |
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".