the most relevant connections according to the
confidence ranking generated by the algorithm.
Finally, a module to identify significant enrich-
ment in GO terms of each (sub-) network was
added to identify molecular functions in which
most of genes are involved.
17.5.4 Example of Multiple Algorithm
Calculation
As an example, we uploaded the list of 1,190 tis-
sue-specific genes identified in Verdier et al.
( 2013 ), we selected the transcriptomic data of the
major plant organs (i.e., leaf, root, nodule, seed,
pod, petiole, andflower) and chose to predict
networks using relevant network (i.e., RN) and
GGM algorithms with default settings
(Fig.17.4a). After submitting thisfirst task, we
merged both network predictions and extracted the
immediate connections of all TFs, options avail-
able in the“Submit query”menu. Figure17.4b
represents the predicted network with nodes (i.e.,
genes) connected by purple lines determined by
RN algorithm and gray lines determined by GGM.
In Fig.17.4c, we chose to visualize similarities
between these two networks by highlighting con-
nections identified by both methods, which pre-
sumably represents a more robust prediction.
17.6 Outlook
In this chapter, we described new tools that are
available to store, visualize, and analyzeL. japo-
nicustranscriptomic data. Most of these tran-
scriptomic data were obtained from the Affymetrix
GeneChip. However, the development of the Ag-
ilent NimbleGen custom gene expression array for
L. japonicus(named 4×44_Kazusa-001 or A-
GEOD-14826) has recently increased the amount
of transcriptomic data. This array contains 44,000
probe sets of 60-mer oligonucleotides, which
correspond to 21,495 genes. This array has already
been used in various experiments, such as cell-
type-specific transcriptome analysis in nodules (E-
GEOD-34753; Takanashi et al. 2012 ), constitutive
over-expression of LjMyb14 (E-GEOD-31739;
Shelton et al. 2012 ), and glutathione-induced
elicitation of defense responses (E-GEOD-31240)
and should provide more data in the next years. In
parallel to microarrays, RNA-seq technology is
rapidly evolving and will, no doubt, produce much
transcriptome data for Lotus in the near future.
All novel transcriptomic data generated for
L. japonicususing different array or RNA-seq
technologies can, in principal, be integrated into
the LjGEA, using recently developed normaliza-
tion strategies such as the one described in Battke
and Nieselt ( 2011 ) to provide a more compre-
hensive understanding of Lotus biology.
Lotus is closely related to major grain legumes
such as lentil, fava, pea, chickpea, common bean,
mung bean, soybean, and pigeon pea. By identi-
fication of putative orthologs of Lotus genes in
the two other model legumes, Medicago and
Soybean, the LjGEA provides a useful tool for
translational genomics, which can be useful for
basic and applied research in many crop legumes.
Moreover, the LegumeGRN Web server contains
a feature for“across-species comparison,”which
takes into account ortholog genes from Medicago
and Soybean. Indeed, regulatory networks gov-
erning basic cellular functions are conserved in
diverse species (Stuart et al. 2003 ). The com-
posite network generated from different species
will display color-coded edges representing the
network connections for each different species.
This multi-species GRN comparison will enable
detection of evolutionary conserved gene regu-
latory (sub-) networks and help to place GRNs in
a phylogenetic context. This feature might be
extended in the future to other legumes as new
genomic data become available.ReferencesAceituno F, Moseyko N, Rhee S, Gutierrez R (2008) The
rules of gene expression in plants: organ identity and
gene body methylation are key factors for regulation
of gene expression inArabidopsis thaliana. BMC
Genom 9(1):438
Asamizu E, Nakamura Y, Sato S, Tabata S (2005)
Comparison of the transcript profiles from the root and
the nodulating root of the model legume Lotus
japonicusby serial analysis of gene expression. Mol
Plant Microbe Interact 18(5):487– 49817 A Tutorial onLotus japonicusTranscriptomic Tools 197