Combining GCN Structural Learning with LLM Chemical Knowledge for Enhanced Virtual Screening
Bibliographic record
Abstract
Virtual screening plays a critical role in modern drug discovery by enabling the identification of promising candidate molecules for experimental validation. Traditional machine learning methods, such as support vector machines (SVM) and XGBoost, rely on predefined molecular representations, often leading to information loss and potential bias. In contrast, deep learning approaches, particularly graph convolutional networks (GCNs), offer a more expressive and unbiased alternative by operating directly on molecular graphs. Meanwhile, large language models (LLMs) have recently demonstrated state-of-the-art performance in drug design thanks to their capacity to capture complex chemical patterns from large-scale data via attention mechanisms. In this paper, we propose a novel hybrid architecture that combines GCNs with LLM-derived embeddings, evaluated on both kinase-related data sets, which are well-established therapeutic targets of high biological significance, and non-kinase data sets such as the glucocorticoid receptor and PPARG, demonstrating the broader applicability of our approach. Our model introduces a layer-wise concatenation strategy, where LLM embeddings are injected after each GCN layer rather than solely at the final layer. This design enables progressive enrichment of the learned molecular representations with global chemical context throughout the network's depth. The LLM embeddings can be precomputed and stored in a molecular feature library, maintaining computational efficiency during training and inference. We conduct a comprehensive comparison against standalone GCN, Molformer, SVM, and XGBoost baselines, demonstrating that our method achieves superior performance, with an accuracy of 88.7%, compared to 86.8% for GCN, 85.1 for molformer, 85.0% for XGBoost, and 84.7% for SVM. These improvements are practically significant in real-world virtual screening scenarios, where even small gains can reduce false positives and accelerate candidate prioritization.
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 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 | 0.002 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".