AXOLOTL: an accurate method for detecting aberrant gene expression in rare diseases using coexpression constraints
Leng, F.; Liu, Y.; Zhang, J.; Shen, Y.; Liu, X.; Wang, Y.; Xu, W.
Show abstract
BackgroundThe assessment of aberrant transcription events in patients with rare diseases holds promise for significantly enhancing the prioritization of causative genes, a practice already widely employed in clinical settings to increase diagnostic accuracy. Nevertheless, the entangled correlation between genes presents a substantial challenge for accurate identification of causal genes in clinical diagnostic scenarios. Currently, none of the existing methods are capable of effectively modeling gene correlation. MethodsWe propose a novel unsupervised method, AXOLOTL, to identify aberrant gene expression events in an RNA expression matrix. AXOLOTL effectively addresses biological confounders by incorporating coexpression constraints. ResultsWe demonstrated the superior performance of AXOLOTL on representative RNA-seq datasets, including those from the GTEx healthy cohort, mitochondrial disease cohort and Collagen VI-related dystrophy cohort. Furthermore, we applied AXOLOTL to real case studies and demonstrated its ability to accurately identify aberrant gene expression and facilitate the prioritization of pathogenic variants.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Accelerate the discovery of genetic variants in mitochondrial diseases with VIOLA: Variant PrIOritization using Latent space 96%
- WEVar: a novel statistical learning framework for predicting noncoding regulatory variants 95%
- Benchmarking copy number aberrations inference tools using single-cell multi-omics datasets 95%
Similar papers in this journal
- MetaRNN: Differentiating Rare Pathogenic and Rare Benign Missense SNVs and InDels Using Deep Learning 94%
- Diagnostic Evidence GAuge of Single cells (DEGAS): A flexible deep-transfer learning framework for prioritizing cells in relation to disease 94%
- scGRNom: a computational pipeline of integrative multi-omics analyses for predicting cell-type disease genes and regulatory networks 94%
Similar papers in this journal
- NIMBus: a Negative Binomial Regression based Integrative Method for Mutation Burden Analysis 95%
- GEOlimma: Differential Expression Analysis and Feature Selection Using Pre-Existing Microarray Data 94%
- Random Walk with Restart on multilayer networks: from node prioritisation to supervised link prediction and beyond 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.