PCA outperforms popular hidden variable inference methods for molecular QTL mapping

Abstract

Estimating and accounting for hidden variables is widely practiced as an important step in quantitative trait locus (QTL) analysis for improving the power of QTL identification. Here we benchmark popular hidden variable inference methods including surrogate variable analysis (SVA), probabilistic estimation of expression residuals (PEER), and hidden covariates with prior (HCP) against principal component analysis (PCA; a well-established dimension reduction and factor discovery method) through comprehensive simulation studies and show that PCA is faster and better-performing. In particular, we show that under realistic simulation scenarios, PEER---the most popular hidden variable inference method for QTL mapping to date---does not have the alleged advantage that its performance does not deteriorate as the number of inferred covariates increases. Additionally, our real data analysis shows that PEER produces identical results as PCA at best, is at least three orders of magnitude slower, and can fail to capture important variance components of the molecular phenotype data in certain cases. Based on these results and the fact that PCA is much easier to interpret and use, we contend that PCA should be the preferred hidden variable inference method for QTL mapping until evidence suggests otherwise. To help researchers use PCA in their QTL analysis, we provide an R package PCAForQTL along with a detailed tutorial.

Rotating U-shaped snap gripping point positioning method based on YOLOv5

“…A part of scholars [4] based on Faster R-CNN [5] algorithm, using two networks to train angle information and position information separately, the former network learns position features, the latter network learns angle features, and finally predicts a five-dimensional grasping box, this method repeatedly learns data and fails to share the bit-pose features. Some other scholars [6] proposed a segmented detection method, which first segmented the target with YOLOv4 and then used subpixel edge extraction and PCA [7] to calculate the angle value with the detected results. However, this method requires autonomous setting of operators and thresholds for angle calculation, resulting in underutilization of the powerful learning capability of deep learning.…”

UNet++ Compression Techniques for Kidney and Cyst Segmentation in Autosomal Dominant Polycystic Kidney Disease

“…Furthermore, PCA-based compression is less susceptible to the influence of initial weight values, enabling consistent results across multi-ple iterations [13]. Additionally, this technique proves effective for networks with heterogeneous structures by identifying the most crucial filters or channels based on their contribution to the overall network performance [14]. Researchers have explored various model compression techniques to reduce the size and computational complexity of deep neural networks.…”

PCA outperforms popular hidden variable inference methods for molecular QTL mapping

“…The R package PCAForQTL and a tutorial on using PCA for hidden variable inference in QTL mapping are available at https://github.com/heatherjzhou/PCAForQTL [ 53 ]. The code used to generate the results in this work is available at https://doi.org/10.5281/zenodo.6788888 [ 54 ]. In addition, this work makes use of the following data and software:…”