Title page for ETD etd-10082010-145739


Document Type Master's Dissertation
Author Coetzer, Nanette
Email nanette.coetzer@gmail.com
URN etd-10082010-145739
Document Title SSHscreen and SSHdb : software for microarray-based screening and sequence management of cDNA libraries
Degree MSc
Department Biochemistry
Supervisor
Advisor Name Title
Prof D K Berger Supervisor
Keywords
  • cDNA libraries
  • Arabidopsis
  • software
  • microarray-based screening
Date 2010-04-28
Availability unrestricted
Abstract
A pipeline was developed for the quantitative screening and sequence management of clones from suppression subtractive hybridization (SSH) cDNA libraries. The pipeline is particularly useful for gene discovery in non-sequenced organisms, and was illustrated with SSH library data from pearl millet (Pennisetum glaucum) and cowpea (Vigna unguiculata) and Arabidopsis (Arabidopsis thaliana) ecotype Kil-0. The objective of each library was to identify stress-response genes.

cDNA microarrays provide a high-throughput screening method. Accordingly, these SSH libraries were amplified by PCR and spotted onto glass microarray slides. Subtracted and un-subtracted cDNA samples, that were used to construct the SSH libraries were prepared as Cy3- and Cy5-labeled targets and hybridized to the microarrays. The R package SSHscreen version 2.0.0, available from http://microarray.up.ac.za/SSHscreen/, was developed to analyze the resulting microarray data using limma (linear models for microarray data) functions. Commonly, loess normalization is used for within-slide normalization, however this is based on the assumption that most of the genes on the array are not differentially expressed. This is legitimate for most whole genome microarray experiments, however it is not appropriate when the array is constructed from an SSH library which is enriched for differentially expressed genes. Therefore, control spot-based normalization was used in the SSHscreen analysis. Empirical Bayes methods were employed to calculate the moderated t-statistic using functions from the limma package. This procedure in effect borrows information from the ensemble of genes to aid with inference about individual genes, taking advantage of the parallel structure whereby the same model is fitted to the data for each gene. In the Arabidopsis, pearl millet and cowpea forward libraries, 18%, 58% and 58% of the clones were identified as significantly up-regulated (adjusted p-value < 0.05) and in the reverse libraries, 18%, 30% and 28% significantly down-regulated, respectively.

SSHscreen analysis was used to assist in selection of clones for sequencing. The SSHscreen data output (ranked gene lists in terms of differential expression), as well as the selected sequences in FASTA format were uploaded to SSHdb. For the Arabidopsis library, 114 out of the 262 sequenced clones (55%) were identified as unique/non-redundant; and for the pearl millet and cowpea libraries respectively, 37% and 33% of the sequenced clones were unique. SSHdb was developed as a web-based tool for sequence management and annotation of clones in SSH libraries and can freely be accessed at http://sshdb.bi.up.ac.za. BLAST analysis that was carried out when sequences were uploaded to SSHdb was used to combine clones with the same sequence into redundant partner groups, as well as identify putative annotations for each group.

Individual clones from the abovementioned SSH libraries were selected and an independent technique, quantitative PCR, was used to validate the microarray/SSHscreen results. The pipeline was applied successfully to Arabidopsis, pearl millet and cowpea SSH cDNA libraries. Interesting genes in each case were identified for further study.

Copyright 2009, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria

Please cite as follows:

Coetzer, N 2009, SSHscreen and SSHdb : software for microarray-based screening and sequence management of cDNA libraries, MSc dissertation, University of Pretoria, Pretoria, viewed yymmdd < http://upetd.up.ac.za/thesis/available/etd-10082010-145739 / >

E10/482/gm

Files
  Filename       Size       Approximate Download Time (Hours:Minutes:Seconds) 
 
 28.8 Modem   56K Modem   ISDN (64 Kb)   ISDN (128 Kb)   Higher-speed Access 
  dissertation.pdf 14.28 Mb 01:06:07 00:34:00 00:29:45 00:14:52 00:01:16

Browse All Available ETDs by ( Author | Department )

If you have more questions or technical problems, please Contact UPeTD.