Genomics:GTL

Genomes to Life Contractor-Grantee Workshop III
February 6-9, 2005, Washington, D.C.

Genomics:GTL Program Projects

Sandia National Laboratories

18

Microarray Analysis using VxInsight and PAM

George S. Davidson*¹ (GSDAVID@sandia.gov), David Hanson², Shawn Martin¹, Margaret Werner-Washburne², and Mark D. Rintoul¹

¹Sandia National Laboratories, Albuquerque, NM and ²University of New Mexico, Albuquerque, NM

In 2001 Hihara et al. [1] published a series of microarray experiments describing gene expression changes in the cyanobacterium Synechocystis sp. PCC 6803 in response to acclimation from a low light level (20 µmol photons m^{-2 sec}-1 to 300 µmol photons m^{-2 sec}-1). These data, which are publicly available from the KEGG database [2], have been reanalyzed using the VxInsight genome tools [3] from Sandia National Laboratories, and PAM [4] from Stanford. The analysis served as a test-bed for preparing the VxInsight tools to work with bacterial microarray data and associated databases. Here we present the results of clustering the individual experiments and the genes by co-expression. Interestingly, a number of experimental design issues are also raised. We examined lists of genes generated by PAM and by VxInsight that include significant differences in expression under the low light (LL) and the high light (HL) experimental conditions. We discuss the methodologies and the visual user interface linking these genes to online annotations and regulatory networks.

Hihara et al. measured total mRNA from HL conditions at 15 min, 1 hr, 6 hr, and 15 hr. These were compared to the mRNA levels measured under LL conditions, which served as control data. Expression levels were measured with CyanoCHIP version 0.8 from TaKaRa. These experiments revealed 84 ORFs with up regulated expression and 80 ORFs with down regulated expression after exposure to HL. Almost all of the photosystem I genes were immediately down regulated, while genes associated with photosystem II showed more complicated patterns. Both observations are consistent with the increasing PSII/PSI ratio in response to HL which involves an initial shift away from of PSI and the gradual construction of PSII (generally completed within 60 min.).

VxInsight found three groups of experiments, as shown in Figure 1, the first of which clearly reflects the stable, ongoing response to HL. The second captures the intermediate state when many of the PSII proteins are being synthesized, or have largely become available as the cells shift toward a higher metabolic plane. The third group contains a random mixture of arrays from time points throughout the experiment, which suggests that technical problems were confounding the measurements (something that is not uncommon in microarray experiments). We identified genes with significantly different expressions between late experimental conditions (first group) and the second group, which consisted of an equal number of measurements made at 15 minutes and at 60 minutes. VxInsight and PAM identified many of the same genes that Hihara et al. found; however the differences offer opportunities for deeper study. We present these genes with their scores and demonstrate the interactive analysis of these lists, including the use of KEGG pathways.

Figure 1. The three clusters of arrays in the Hihara et al. experiment (top left), together with the gene clusters (top right). The top 15 genes relevant to the shift from HL to LL are listed on the bottom left, where the list contains links to KEGG networks, as shown on the bottom right.

References

1. Hihara, Y., et al., DNA microarray analysis of cyanobacterial gene expression during acclimation to high light. The Plant Cell, 2001. 13: p. 793-806.
2. Hihara, Y., http://www.genome.jp/kegg/expression/.
3. Davidson, G.S., et al., High throughput instruments, methods, and informatics for systems biology (also to appear as: Robust Methods for Microarray Analysis, in Genomics and Proteomics Engineering in Medicine and Biology, IEEE/Wiley Press, Metin Akay, editor, in press). 2003, Sandia National Laboratories SAND2003-4664: Albuquerque, New Mexico 87185.
4. Tibshirani, R., et al., Diagnosis of multiple cancer types by shrunken centroids of gene expression. PNAS, 2002. 99(10): p. 6567-6572.

Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy’s National Nuclear Security Administration under contract DE-AC04-94AL85000. This work was supported by the U.S.Department of Energy’s Genomics: GTL program (genomicsgtl.energy.gov) under project, “Carbon Sequestration in Synechococcus sp.: From Molecular Machines to Hierarchical Modeling,”(www.genomes-to-life.org).

* Presenting author