M-quantile Regression Analysis of Temporal Gene Expression Data

Abstract

In this paper, we explore the use of M-regression and M-quantile coefficients to detect statistical differences between temporal curves that belong to different experimental conditions. In particular, we consider the application of temporal gene expression data. Here, the aim is to detect genes whose temporal expression is significantly different across a number of biological conditions. We present a new method to approach this problem. Firstly, the temporal profiles of the genes are modelled by a parametric M-quantile regression model. This model is particularly appealing to small-sample gene expression data, as it is very robust against outliers and it does not make any assumption on the error distribution. Secondly, we further increase the robustness of the method by summarising the M-quantile regression models for a large range of quantile values into an M-quantile coefficient. Finally, we employ a Hotelling T 2 -test to detect significant differences of the temporal M-quantile profiles across conditions. Simulated data shows the increased robustness of M-quantile regression methods over standard regression methods. We conclude by using the method to detect differentially expressed genes from time-course microarray data on muscular dystrophy.

Regresión cuantil para predicción de caracteres complejos en bovinos Suizo Europeo usando marcadores SNP y pedigrí

“…El uso de un enfoque sólido como QR hace que la inferencia sea menos sesgada y esté menos sujeta a falsos positivos (2) . En estudios recientes que utilizan QR, se describen aplicaciones diversas como estudios de asociación genética (4) , genética de poblaciones (5) , expresión génica (6,7) y selección genómica (8)(9)(10) . Uno de los primeros estudios donde se utilizó QR para predecir el mérito genético individual lo presentaron Nascimento et al (11) , quienes utilizaron datos simulados encontrando ventajas al usar QR frente a metodologías convencionales.…”

Scale estimation and data-driven tuning constant selection for M-quantile regression

“…Similarly to SAE methods, MQ regression models have been used in poverty mapping (Tzavidis et al, 2008;Giusti et al, 2009). MQs were also used in estimating acidity in north-eastern US lakes (Pratesi et al, 2008), and in an analysis of temporal gene expression data (Vinciotti and Keming, 2009). Also, MQ random-effects models were introduced by Tzavidis et al (2016) and Borgoni et al (2018) with applications to longitudinal data.…”