Abstract 6375: Augmented intelligence to define drug targets associated with triple negative breast cancer (TNBC) metabolic reprogramming
Bibliographic record
Abstract
Abstract Introduction: Basal-like breast cancer (BLBC; TNBC) cells use aerobic glycolysis at a higher rate than Luminal A (LumA) cells. Metabolic reprogramming using aerobic glycolysis (Warburg effect), is correlated with increased aggressiveness of cancer cells and poor outcome in patients. Therefore, genes involved in this pathway are promising targets for developing cancer therapeutics. Method: MIMs has developed a unique platform of augmented intelligence that combines bioinformatics, systems biology and artificial intelligence. It integrates multi-layered omics data with a knowledge-base that aggregates structured data from more than 140 databases as well as unstructured data from the scientific literature. Using this approach a genetic interaction graph (GI-Graph) is inferred per patient, capturing functional relationships between genes in the specific context of the tumor. The GI-Graphs are subsequently used to train supervised machine learning algorithms for predicting gene functionality and potential as a target. Preliminary analysis was performed on transcriptomic data from 321 LumA and 162 TNBC samples, from the TCGA Data Portal and a predictive model was developed using two GI-Graphs from the two subgroups. Briefly, subgraphs, containing genes with functional interactions with the known genes in OXPHOS and glycolysis pathways, were used to extract gene attributes, and to build an algorithm that predicts involvement of a gene in the Warburg effect. Using a testing gene set extracted from the literature the performance of the model was assessed, which showed a true positive rate of 18% and a false positive rate of 0.36%, and outperformed 5-times the classical bioinformatics tools. Results: This model predicted 108 genes as the top 1% genes being involved in the metabolic reprogramming of TNBC. Additional information from MIMs’ platform, including differential gene expression between LumA and TNBC, gene pleiotropy and essentiality and the topological metrics, enabled the life scientists to further refine the gene lists, based on the expected characteristics of a good target in oncology. Following this process, 30 genes were selected as potential targets controlling the metabolic reprogramming of TNBC, among which 4 genes were already evaluated in TNBC clinical trials. Conclusion: Our preliminary data strongly supports that the predictive model based on the GI-Graphs has the potential to identify promising therapeutic targets for TNBC. Citation Format: Sarah Jenna, Benjamin Boucher, Liebaut Dudragne, Abdoulaye Baniré Diallo. Augmented intelligence to define drug targets associated with triple negative breast cancer (TNBC) metabolic reprogramming [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6375.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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 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".