Using Auto-Encoders to Create Encodings for Three-Dimensional Protein Structure Information
Bibliographic record
Abstract
Proteins play a crucial role in various biological processes, serving as the building blocks and machines of life. Therefore, understanding their structure and function is paramount for advancing our knowledge of structural biology. The Protein Data Bank (PDB) [3] files have been an integral part of helping researchers decipher the complex workings of proteins. PDB files provide three-dimensional Cartesian coordinates of protein structures which are used as a stepping stone for other protein structure tools, such as protein classification. Efficient protein classification is vital for organizing and categorizing the large number of proteins discovered to date. It enables researchers to identify functional relationships, predict protein functions, and gain insights into their evolutionary history. However, current protein structural classification systems like CATH [22], SCOP [2] and SCOPe [5] have some limitations, such as complicated protein structure domain descriptions, manual and subjective classification, lack of customizability for users with different classification needs, and handling the increasing volume of protein structure data. Recently, image processing has advanced significantly, mainly due to neural networks such as Convolutional Neural Networks (CNNs) and auto-encoders. This work aims to harness the remarkable success of learned representations and CNNs for image processing by proposing a foundation model for the development of new encodings from three-dimensional protein structure information for various classification needs. The protein encodings will be helpful in other protein structure-related problems such as protein structure prediction, protein function prediction, and drug discovery.
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 | 0.000 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".