The software listed in this site is for non-commercial use only. It is provided as is, and it comes with no guaranttee. Neither current and former members of Glazebrook and Katagiri Labs nor Univ. of Minnesota are responsible for any damage caused by use of the software.

• Designing PCR primer sets for Arabidopsis T-DNA insertion line genotyping



• Calculating expression values from probe fluorescence values in the miniarray
  
  The miniarray has symmetric and circularly overlapping probe replicate design. This script converts the fluorescence values from probe spots to expression level values for each probe. This script includes SBQ (see below). See Sato et al. (2007).



• SBQ (Stable gene-Based Quantile normalization)
  
  In the miniarray, 107 genes that have relatively stable, diverse expression levels are used as between-arrays normalization genes. SBQ uses expression levels of these genes to perform between-arrays normalization. See Sato et al. (2007).



• LEGG (Locally linear Embedding Graph Generator)
  
  LEGG uses the concept of Locally Linear Embedding to perform non-linear dimensionality reduction. This local dimensionality information is converted into a graph to show the relationships among the data points. LEGG uses the same mathematical principle as LCF (Local Context Finder; Katagiri and Glazebrook (2003)), but LEGG uses a different algorithm which is faster and more accurate. See van Poecke et al. (2007).



• RSPR-NR (Row-specific, Sorted PRincipal component-guided method for Noise Reduction)
  RSPR-NR statistically reduces random noise in large data set. Unlike PCA, RSPR-NR retains samll feature signals in data very well.
  See Foley, J. and Katagiri, F. (2009).
  



• Soon to come! RepEdLEGG (Repetitive Euclidean distance-based Locally linear Embedding Graph Generator). EdLEGG runs LEGG algorithm using a linear model. By repeating the EdLEGG procedure on the residual of the first EdLEGG procedure, RepEdLEGG finds weak but significant and non-reduandant similarities in a data set, as well as strong and non-redundant similarities.