in other proteins. If you hover over the red exclamation points,
the genes that share this peptide will be listed. If you search
these other genes, you will find that they are HSP90 paralogs.
Mechanistically, this should make sense because paralogs typically have highly similar genic sequences.
6. To export the data, click on “Export results” at the top of the
page. A TSV file containing information about PTMs of
HSP90.1 should be downloaded automatically.
3.10 Protein–Protein
Interaction Networks
There are several databases to explore for Arabidopsis protein-protein interactions. A big Arabidopsis-specific one is the BAR’s new
Arabidopsis Interactions Viewer 2 (AIV2). However, it is advisable
to examine other databases, such as IntAct at http://www.ebi.ac.
uk/intact/ [63] or BioGRID at http://thebiogrid.org [64], both
of which are not specific for Arabidopsis, or AtPID (http://www.
megabionet.org/atpid/) by Li et al. [65], as literature curation
efforts are by no means complete for any of these databases. Note
that as of a couple of years ago, the AIV2 described in the next
section is able to automatically query PSICQUIC-enabled databases for other Arabidopsis interactions, thereby facilitating the
searching of multiple databases (PSICQUIC is an effort to standardize the access to molecular interaction databases, see https://
github.com/PSICQUIC/).
3.10.1 Arabidopsis
Interactions Viewer 2 (AIV2)
The BAR’s Arabidopsis Interactions Viewer 2 at http://bar.
utoronto.ca/interactions2/ [66] currently permits the exploration
of 80,009 predicted and 62,626 experimentally determined protein-protein interactions curated by BIND, the BAR, IntAct,
TAIR, etc., along with ~2.8M protein-DNA interactions (PDIs).
One may submit a list of gene (product) identifiers and the AIV will
return the interactors of the proteins. It is possible to return only
experimentally documented interactions, or all interactions including those predicted through the use of the interolog method
(interacting ortholog) described in Geisler-Lee et al. [67] or via
docking [66]. Attractive features of the AIV include the ability to
upload Cytoscape files (.cys files) as well as the ability to color nodes
by their expression level in different tissues to help define subnetworks in different tissue types. Unlike ePlant, a nice “layered” view
from the outside of the cell to the inside of the cell is available, and,
for PDIs, a matrix of interactions is available to help ascertain which
DNA targets have transcription factors binding in common.
1. Go to http://bar.utoronto.ca/interactions2/.
2. Enter an AGI identifier, or a list of identifiers, and select any of
the options you wish. The default setting will return all experimentally determined and predicted interactions for your gene
products of interest. For this example we will not check any of
Arabidopsis Bioinformatics
69
the genes that share this peptide will be listed. If you search
these other genes, you will find that they are HSP90 paralogs.
Mechanistically, this should make sense because paralogs typically have highly similar genic sequences.
6. To export the data, click on “Export results” at the top of the
page. A TSV file containing information about PTMs of
HSP90.1 should be downloaded automatically.
3.10 Protein–Protein
Interaction Networks
There are several databases to explore for Arabidopsis protein-protein interactions. A big Arabidopsis-specific one is the BAR’s new
Arabidopsis Interactions Viewer 2 (AIV2). However, it is advisable
to examine other databases, such as IntAct at http://www.ebi.ac.
uk/intact/ [63] or BioGRID at http://thebiogrid.org [64], both
of which are not specific for Arabidopsis, or AtPID (http://www.
megabionet.org/atpid/) by Li et al. [65], as literature curation
efforts are by no means complete for any of these databases. Note
that as of a couple of years ago, the AIV2 described in the next
section is able to automatically query PSICQUIC-enabled databases for other Arabidopsis interactions, thereby facilitating the
searching of multiple databases (PSICQUIC is an effort to standardize the access to molecular interaction databases, see https://
github.com/PSICQUIC/).
3.10.1 Arabidopsis
Interactions Viewer 2 (AIV2)
The BAR’s Arabidopsis Interactions Viewer 2 at http://bar.
utoronto.ca/interactions2/ [66] currently permits the exploration
of 80,009 predicted and 62,626 experimentally determined protein-protein interactions curated by BIND, the BAR, IntAct,
TAIR, etc., along with ~2.8M protein-DNA interactions (PDIs).
One may submit a list of gene (product) identifiers and the AIV will
return the interactors of the proteins. It is possible to return only
experimentally documented interactions, or all interactions including those predicted through the use of the interolog method
(interacting ortholog) described in Geisler-Lee et al. [67] or via
docking [66]. Attractive features of the AIV include the ability to
upload Cytoscape files (.cys files) as well as the ability to color nodes
by their expression level in different tissues to help define subnetworks in different tissue types. Unlike ePlant, a nice “layered” view
from the outside of the cell to the inside of the cell is available, and,
for PDIs, a matrix of interactions is available to help ascertain which
DNA targets have transcription factors binding in common.
1. Go to http://bar.utoronto.ca/interactions2/.
2. Enter an AGI identifier, or a list of identifiers, and select any of
the options you wish. The default setting will return all experimentally determined and predicted interactions for your gene
products of interest. For this example we will not check any of
Arabidopsis Bioinformatics
69