QTL EM Algorithm Mapping and Hotspots Detection
Description
For QTL mapping, this package comprises several functions designed to execute diverse
tasks, such as simulating or analyzing data, calculating significance thresholds, and
visualizing QTL mapping results. The single-QTL or multiple-QTL method, which enables
the fitting and comparison of various statistical models, is employed to analyze the
data for estimating QTL parameters. The models encompass linear regression, permutation
tests, normal mixture models, and truncated normal mixture models. The Gaussian stochastic
process is utilized to compute significance thresholds for QTL detection on a genetic
linkage map within experimental populations. Two types of data, complete genotyping, and
selective genotyping data from various experimental populations, including backcross, F2,
recombinant inbred (RI) populations, and advanced intercrossed (AI) populations, are
considered in the QTL mapping analysis. For QTL hotspot detection, statistical methods can
be developed based on either utilizing individual-level data or summarized data. We have
proposed a statistical framework capable of handling both individual-level data and
summarized QTL data for QTL hotspot detection. Our statistical framework can overcome the
underestimation of thresholds resulting from ignoring the correlation structure among
traits. Additionally, it can identify different types of hotspots with minimal computational
cost during the detection process. Here, we endeavor to furnish the R codes for our QTL
mapping and hotspot detection methods, intended for general use in genes, genomics, and
genetics studies. The QTL mapping methods for the complete and selective genotyping designs
are based on the multiple interval mapping (MIM) model proposed by Kao, C.-H. , Z.-B. Zeng
and R. D. Teasdale (1999) and H.-I Lee, H.-A. Ho and
C.-H. Kao (2014) , respectively. The QTL hotspot detection
analysis is based on the method by Wu, P.-Y., M.-.H. Yang, and C.-H. Kao (2021)
.