bioRxiv Science⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/533406

DeepTRIAGE: Interpretable and Individualised BiomarkerScores using Attention Mechanism for the Classification of Breast Cancer Sub-types

Beykikhoshk, A.·Quinn, T. P.·Lee, S. C.·Tran, T.·Venkatesh, S.
View original source ↗Export metadata ↓

Abstract

MotivationBreast cancer is a collection of multiple tissue pathologies, each with a distinct molecular signature that correlates with patient prognosis and response to therapy. Accurately differentiating between breast cancer sub-types is an important part of clinical decision-making. Already, this problem has been addressed using machine learning methods that separate tissue samples into distinct groups. However, there remains unexplained heterogeneity within the established sub-types that cannot be resolved by the commonly used classification algorithms. In this paper, we propose a novel deep learning architecture, called DeepTRIAGE (Deep learning for the TRactable Individualised Analysis of Gene Expression), which not only classifies cancer sub-types with comparable accuracy, but simultaneously assigns each patient their own set of interpretable and individualised biomarker scores. These personalised scores describe how important each feature is in the classification of each patient, and can be analysed post-hoc to generate new hypotheses about intra-class heterogeneity. ResultsWe apply the DeepTRIAGE framework to classify the gene expression signatures of luminal A and luminal B breast cancer sub-types, and illustrate its use for genes and gene set (i.e., GO and KEGG) features. Using DeepTRIAGE, we find that the GINS1 gene and the kinetochore organisation GO term are the most important features for luminal sub-type classification. Through classification, DeepTRIAGE simultaneously reveals heterogeneity within the luminal A biomarker scores that significantly associate with tumour stage, placing all luminal samples along a continuum of severity. Availability and implementationThe proposed model is implemented in Python using Py-Torch framework. The analysis is done in Python and R. All Methods and models are freely available from https://github.com/adham/BiomarkerAttend.

Source connections

Explore related subjects

biophysicsPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Beykikhoshk, A., Quinn, T. P., Lee, S. C., Tran, T., Venkatesh, S.. 2019-01-29. DeepTRIAGE: Interpretable and Individualised BiomarkerScores using Attention Mechanism for the Classification of Breast Cancer Sub-types. https://doi.org/10.1101/533406

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Autonomous Homeostatic Synthetic Cells via Self-Gating DNA Nanopores

Homeostasis is a fundamental hallmark of living organisms, arising from the complex interplay between biochemical reactions and regulatory feedback systems. Reconstituting such self-regulating behaviour in minimal synthetic cells enables continuous, persistent operation of biochemical reactions for extended amount of time. In this work, we demonstrate a minimal homeostatic synthetic cell capable of autonomous flux regulation using DNA nanotechnology and bottom-up synthetic biology. Our homeostatic architecture consists of Giant Unilamellar Vesicles (GUVs) equipped with gated DNA nanopores, encapsulated in vitro transcription (IVT) machinery, and an RNA degradation system. We achieve homeostasis under varying external chemical stimuli specifically varying concentrations of rNTPs by implementing a negative feedback loop between rNTP influx and RNA production. In our system, DNA nanopores facilitate the influx of rNTPs from the external environment, driving internal transcription. Crucially, the transcription process generates RNA "blockers" designed to bind and gate the DNA nanopores, thereby attenuating further rNTP influx. Our system is dynamic as encapsulated RNases slowly degrade the RNA blockers, allowing the pores to reopen as blocker concentration goes down. We first characterise the functionality and gating efficiency of the DNA nanopores using both pre-synthesised and in situ produced DNA and RNA blockers. We then demonstrate that rNTP flux through these pores is sufficient to drive IVT within the GUVs. Finally, by integrating these modules, we demonstrate robust homeostasis: the system maintains a steady-state level of RNA production for up to 16 hours. By harnessing the controllability of negative feedback loop, we demonstrate thresholding of the homeostasis level using single-stranded regulator DNA. This work establishes a versatile framework for engineering adaptive and self-sustaining responsive nanomaterials and synthetic cell chassis.

biophysics↗
bioRxiv2026

A Generic Numbering Scheme for TMEM16 Scramblases

The TMEM16 family of calcium-activated phospholipid scramblases (CaPLSs) and chloride channels (CaCCs) performs diverse physiological functions that include regulation of blood coagulation and apoptotic signaling, through a shared ten-transmembrane-helix (TM) architecture organized around a hydrophilic lipid-translocating groove. Mechanistic studies of TMEM16 family members have been hampered by the absence of a unified positional reference framework that would permit direct comparison of structurally equivalent residues across paralogs with different sequence numbering systems. Here we introduce a generic numbering scheme for TMEM16 scramblases (GNS-TMEM16), modeled on the Ballesteros & Weinstein system established for class A G protein-coupled receptors. A reference alignment (TMEM16-RA) was constructed from twelve human and mouse TMEM16 scramblases (TMEM16C/D/E/F/G/J) using structure-based ClustalW alignment of the ten TM helices. From this alignment, a TM-specific reference residue (TsRR) was identified for each helix by hierarchical application of three criteria: (1) 100% conservation in the core TMEM16-RA; (2) conservation in an augmented reference alignment (TMEM16-ARA) incorporating a group of phylogenetically more distant homologs composed of nhTMEM16, afTMEM16, TMEM16K, TMEM16A, and TMEM16B; and (3) structural and functional considerations, including helix-perturbing character, groove localization, conserved motif membership, and central TM position. The resulting ten TsRRs are Y1.50, W2.50, R3.50, E4.50, F5.50, P6.50, E7.50, D8.50, W9.50, and E10.50, and are illustrated in mTMEM16F. Each residue is assigned the identifier N.m(k), where N is the TM number, m is the position relative to the TsRR (for which m = 50), and k is the absolute sequence number. Loop residues receive dual identifiers referenced to the TsRRs of both flanking helices. Application of the GNS-TMEM16 is illustrated with the comparisons of the groove-opening measurements using pairwise distances between residues identified by their N.m indices to be corresponding across mTMEM16F, afTMEM16, and nhTMEM16. The results bring to light the advantages of corresponding residues identification in different TMEM16 proteins and show that the mammalian scramblase undergoes substantially larger separation at the extracellular groove entrance than either fungal homolog. Comparison of mutagenesis data guided by N.m correspondence shows at the conserved (E3.55,R6.26) salt-bridge locus, Ala substitution reduces activity more than 100-fold in nhTMEM16 but less than 2-fold in afTMEM16, illustrating that the GNS identifies structural equivalence of position without implying functional equivalence of the residue, which is a distinct advantage of GNS in providing mechanistic interpretation across paralogs. Also described is a protocol for extending the GNS-TMEM16 to uncharacterized protein sequences, including AlphaFold-predicted models, using structural superposition to mTMEM16F. Thus, the presented GNS-TMEM16 provides a stable positional reference for the integration and comparative analysis of structural, computational, and functional data across the TMEM16 family, utilizing a construction strategy applicable to yet other polytopic membrane protein families sharing a common transmembrane fold.

biophysics↗
bioRxiv2026

An agent-based 3D model of non-genetic adaptation in cancer tissues under electrical, mechanical, and hypoxic stress

Non-genetic adaptation enables cancer cells to alter their phenotype under stress without requiring new mutations. However, the mechanisms by which electrical, mechanical, and hypoxic cues combine to shape this process in 3D tissues remain poorly understood. This work presents an agent-based tumor model that integrates vascular oxygen supply, a globally imposed electric field, mechanically mediated crowding and compression cues, phenotype transitions, cell growth, mitosis, death, and inheritance of adaptive memory across division. The simulated tumors exhibit a three-stage trajectory consisting of necrosis onset, transient collapse of live mass, and partial regrowth accompanied by progressive accumulation of adapted cells. Continuous electrical stimulation produces a dose-dependent reduction in live mass while markedly increasing the adapted fraction, with comparatively limited changes in final necrotic burden. This response is strongly conditioned by mechanics and reshapes (and is reshaped by) adaptive capacity. Pulsed stimulation further shows that, in the model, electric field amplitude and temporal schedule jointly determine memory phenomena, phenotypic diversification, and growth recovery. These results show that coupling local oxygen availability, mechanical constraints, electrical forcing, and history-dependent phenotype transitions can generate distinct tissue-level patterns of phenotypic heterogeneity. Both stimulus magnitude and temporal protocol influenced the resulting population structure, suggesting that the history of physical stress may be an important determinant of adaptive dynamics in spatially organized tumor models.

biophysics↗