Abstract B020: A copy number and expression based classifier for breast cancer tumors
Bibliographic record
Abstract
Abstract In our recent study, the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) proposed a new classification of breast cancer tumors into 10 new subtypes based on an integrated analysis of copy number and gene expression of 1,000 genes on 997 samples. These integrative clusters (IntClust) showed different copy number and expression profiles with distinct prognosis. Some of these IntClust were novel to the breast cancer literature, including a high risk ER+ group that have an 11q13/14 amplification (IntClust2) and a favourable prognosis group with no copy number alterations (IntClust4). This unsupervised analysis was validated with another 983 tumors, showing that the 10 groups were reproducible. Here we present a robust multi-platform classifier that predicts the integrative cluster of any breast tumor using data from any existing microarray/sequencing platform. Amongst its main features are the ability to classify each sample independently, -that is the assignment of each specific sample does not depend on the rest of the set of the tumors that are being classified- and its good performance on different microarray and sequencing platforms tested. It also provides the probability for every sample to belong to each of the ten subgroups. We will discuss the challenges encountered when building a proper classifier, such as robust multi-platform feature selection –that is, selecting and adapting the original probes used in the paper to a general setting- and normalization of the data. Copy number features are easy to adapt because they are based on log-ratios against a normal reference; something available in most platforms, but expression features require a more rigorous processing approach to ensure that their distribution is comparable to the original discovery dataset used to derive the ten integrative clusters. Finally, we present results on several publicly available breast cancer datasets that reflect the heterogeneity present in this disease and pinpoint the distinct genomic features that these ten subgroups possess. We have also classified several widely used breast cell lines into the integrative clusters, which can be very helpful in designing functional experiments oriented towards understanding biological processes that drive tumorigenesis in each of these ten new subgroups. The classifier will be available to the scientific community as an R/Bioconductor package. Citation Format: Oscar M. Rueda, Raza H. Ali, Christina Curtis, Suet-Feung Chin, Alejandra Bruna, Roslin Russell, Bernard Pereira, group METABRIC, Samuel Aparicio, Carlos Caldas. A copy number and expression based classifier for breast cancer tumors. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Breast Cancer Research: Genetics, Biology, and Clinical Applications; Oct 3-6, 2013; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2013;11(10 Suppl):Abstract nr B020.
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 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".