Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/659078

Improved detection of gene fusions by applying statistical methods reveals new oncogenic RNA cancer drivers

Dehghannasiri, R.·Freeman, D. E.·Jordanski, M.·Hsieh, G. L.·Damljanovic, A.·Lehnert, E.·Salzman, J.
View original source ↗Export metadata ↓

Abstract

Short AbstractThe extent to which gene fusions function as drivers of cancer remains a critical open question. Current algorithms do not sufficiently identify false-positive fusions arising during library preparation, sequencing, and alignment. Here, we introduce a new algorithm, DEEPEST, that uses statistical modeling to minimize false-positives while increasing the sensitivity of fusion detection. In 9,946 tumor RNA-sequencing datasets from The Cancer Genome Atlas (TCGA) across 33 tumor types, DEEPEST identifies 31,007 fusions, 30% more than identified by other methods, while calling ten-fold fewer false-positive fusions in non-transformed human tissues. We leverage the increased precision of DEEPEST to discover new cancer biology. For example, 888 new candidate oncogenes are identified based on over-representation in DEEPEST-Fusion calls, and 1,078 previously unreported fusions involving long intergenic noncoding RNAs partners, demonstrating a previously unappreciated prevalence and potential for function. Specific protein domains are enriched in DEEPEST calls, demonstrating a global selection for fusion functionality: kinase domains are nearly 2-fold more enriched in DEEPEST calls than expected by chance, as are domains involved in (anaerobic) metabolism and DNA binding. DEEPEST also reveals a high enrichment for fusions involving known and novel oncogenes in diseases including ovarian cancer, which has had minimal treatment advances in recent decades, finding that more than 50% of tumors harbor gene fusions predicted to be oncogenic. The statistical algorithms, population-level analytic framework, and the biological conclusions of DEEPEST call for increased attention to gene fusions as drivers of cancer and for future research into using fusions for targeted therapy.\n\nSignificanceGene fusions are tumor-specific genomic aberrations and are among the most powerful biomarkers and drug targets in translational cancer biology. The advent of RNA-Seq technologies over the past decade has provided a unique opportunity for detecting novel fusions via deploying computational algorithms on public sequencing databases. Yet, precise fusion detection algorithms are still out of reach. We develop DEEPEST, a highly specific and efficient statistical pipeline specially designed for mining massive sequencing databases, and apply it to all 33 tumor types and 10,500 samples in The Cancer Genome Atlas database. We systematically profile the landscape of detected fusions via employing classic statistical models and identify several signatures of selection for fusions in tumors.\n\nSoftware availabilityDEEPEST-Fusion workflow with a detailed readme file is available as a Github repository: https://github.com/salzmanlab/DEEPEST-Fusion. In addition to the main workflow, which is based on CWL, example input and batch scripts (for job submission on local clusters), and codes for building the SBT files and SBT querying are provided in the repository. All custom scripts used for systematic analysis of fusions are also available in the same repository.

Source connections

Explore related subjects

bioinformaticsPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Dehghannasiri, R., Freeman, D. E., Jordanski, M., Hsieh, G. L., Damljanovic, A., Lehnert, E., Salzman, J.. 2019-06-03. Improved detection of gene fusions by applying statistical methods reveals new oncogenic RNA cancer drivers. https://doi.org/10.1101/659078

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

KEEP EXPLORING

Related preprints

bioRxiv2026

Deducing splicing events from transcript sets with splicelogic

Long-read RNA sequencing captures a greater proportion of full-length transcripts, enabling more accurate isoform identification, discovery of novel isoforms, and more precise differential transcript usage analysis. However, comparing transcripts, whether between conditions or between sets such as novel and annotated isoforms, yields complex isoform changes that are hard to manipulate, summarize, and visualize. This is compounded by vertical integration of some tools, where downstream analysis depends on the specific output of a particular upstream tool. To address this gap, we have developed splicelogic, an R/Bioconductor package that makes transcript-set comparisons biologically interpretable by turning them into discrete splicing events, such as exon skipping, alternative splice sites, intron retention, and alternative transcription start and end sites. We designed splicelogic to integrate easily with existing RNA-seq workflows and tools. The splicelogic functions are modular, readable, and extensible, making it easy for users to adapt core logic to their own custom scientific questions or novel splicing definitions. By obtaining interpretable splicing events from transcript changes, splicelogic enables downstream analyses, including annotation of RNA-binding protein recognition sites, alongside summary statistics and visualization. The splicelogic package is publicly available and distributed through Bioconductor.

bioinformatics↗
bioRxiv2026

Contrastive Learning Improves Generalization on M2OR Olfactory Binding Dataset

Olfaction-based biosensors stand to have profound impacts in fields such as medical diagnostics, national security, and drug discovery. A promising mode of vapor phase biodetection mimics volatile organic compound-olfactory receptor (VOC-OR) interactions with target-specific protein or peptide probes; however, lack of experimental coverage remains a core bottleneck. Current computational approaches aim to improve prediction of VOC-OR binding to reduce experimental burden, and current machine learning approaches fail to demonstrate a level of generalization that addresses current engineering challenges. Here, using the Molecular to Olfactory Dataset (M2OR) and Insect Olfactory Binding Protein Database (iOBPdb) datasets, data partitioning methods and their performance are evaluated on a neural network and contrastive learning framework on such VOC-OR binding datasets, varying: model training, architecture, and structure of the input data distribution. It was found that holding out certain chemical or proteomic motifs from the training data diminished model performance with a two layer neural network model. The model was able to generalize to state of the art performance (0.88 AUCROC) on every data split by applying a contrastive learning objective defined by shared molecular constituents and a stochastic sampler. This suggests the contrastive objective provides a strong inductive bias for neural networks to predict novel VOC-OR pairs. These findings highlight that model performance on pairwise biological interaction datasets can be inflated through identity leakage from constituent overlap, and emphasize the need for more robust evaluation protocols. Furthermore, we show contrastive regularization mitigates dataset degeneracy through enforcement of compatibility in the models latent geometry that may extend to other binder target-datasets where identity leakage confounds evaluation.

bioinformatics↗
bioRxiv2026

Plate design confounds donor variation and model evaluation in single-cell perturbation screens

Single-cell screens of immune cells from many donors are used to study why individuals respond differently and to benchmark response models. Each donor-perturbation combination typically occupies one well, controls are grouped, and donors share one layout, so technical effects can look biological. Using a 90-cytokine screen, two compound screens, four cohorts and simulations, we show that well effects explain most replicate-control variation in abundant cell types, controls carry a shared offset, and identical control wells differ by hundreds of differentially expressed genes at a standard threshold, about as many as two receptor-negative proteins do. Negative controls, design-aware estimators and perturbation-specific evaluation change the interpretation of four recent analyses. A deep-learning model that scores below a perturbation-agnostic predictor under standard evaluation outperforms a simple baseline on perturbation-specific differences in two of three cell types (significantly in one with stricter controls), but predicts less well the component shared with inert wells.

bioinformatics↗