target locus

Searching for a target locus in AURA is done by simply typing the search term in the search box: this results in the UTR Searcher page being displayed.

All genes whose HGNC gene symbol, synonyms or UCSC knownGene transcript ID contains the search term are shown along with a description of their function and number of related UTRs. By clicking on a gene panel, the user can display the UTR exon/intron structures and the list of all UTRs of the gene. Transcript names follow the UCSC knownGene naming system.

UTRs can be dragged and dropped into the Select UTRs panel on the right side of the page. By clicking on Explore UTRs, the user eventually proceeds to the UTR view results for all selected UTRs.

trans-factor

Typing the trans-factor name in the search box results in a disambiguation page where all trans-factors whose name or synonyms match the search term are shown. If there is an exact match, the trans-factor targets list is instead shown directly. The user can choose the UTR type, order trans-factor targets by Gene Ontology categories or chromosome and filter them by technique. Eventually, the expression filter allows to select one of the 48 available gene expression profiles: this will cause the display of a subset of all regulatory interactions, namely the ones for which both the trans-factor and the target genes are expressed in that particular tissue or cell line.

When the results are shown, the user can look for UTRs of interest in the various categories and drag and drop selected ones into the selected UTR panel. A textual format card (the Regulatory Factor Card), containing basic annotation and targets data for the factor, can be downloaded from this page.

Eventually, the annotations related to selected UTR can then be explored in the UTR view panel for each UTR.

co-regulation

This search mode allows to execute the “trans-factor” query on multiple query trans-factors and to obtain annotated UTRs which are co-regulated by all the trans-factors in the query. If a specific tissue/cell line is selected in the expression filter, results will include only those trans-factors and targets which are expressed in that tissue/cell line.

It works exactly in the same way as the trans-factor search, both in terms of parameters and of results pages displayed to the user.

sequence

By typing, in plain text format (no FASTA header required) the desired query sequence in the search box, the results page is displayed where you may scroll 5’ and 3’ UTRs matching your query sequence. All matches are displayed with position, score and statistical significance as computed by the BLAST software.

Users can then, by clicking on the buttons at the top of the page, proceed to browse all matching UTRs (through the UTR view as for the other search modes) or download all results in a textual format.

Batch mode

This query mode allows to execute queries on multiple genes at once, and provides multiple options:

• "browser display" permits to obtain the list of UTRs belonging to query genes and to browse them as in the “target locus" search.
• “post-transcriptional regulatory network” allows to build a network with the trans-factor-UTR interactions stored in AURA and involving query genes
• “regulatory element enrichment” allows to compute statistical significance values for the enrichment of regulatory sites of a given trans-factor or cis-elements in query genes UTRs.

batch mode > Browser display

By this option the user can execute the “target locus” query on multiple query genes simultaneously. In order to do so, it is sufficient to type the query genes symbols in the search box, one per line.

The UTR searcher page will then be displayed as in “target locus” search mode.

batch mode > Post-transcriptional regulatory network

By this option the user can visualize the network of direct trans-factor-UTR binding interactions which involve the UTRs associated with your query genes.

In order to perform the query, it is sufficient to type query genes in the search box, one per line. The expression filter also allows to select one of the 48 available gene expression profiles: this will cause the display of a subset of all regulatory interactions, namely the ones for which both the regulator and the target gene are expressed in that particular tissue or cell line. The network will then be shown through an interactive CytoscapeWeb plugin. Node shape varies depending on node type: rounded rectangle for query genes, ellipse for genes interacting with query genes and diamond for cis-elements. Node size increases with the number of genes regulated by the factor represented by the node. On the other hand, network edges are weighted by the number of binding sites of the regulator (source node) on the regulated gene UTRs (destination node): this is displayed by increasing edge thickness.

The network can eventually be downloaded in PDF or GraphML formats by clicking on buttons at the bottom of the page.

batch mode > Regulatory element enrichment

By this option the user can explore which cis-regulatory elements are overrepresented in the query UTRs or which trans-factors yield overrepresented binding sites in the query UTRs.

In order to perform the query, it is sufficient to type query genes in the search box, one per line. The expression filter allows to select one of the 48 available gene expression profiles: this will cause the computation of the enrichment p-value for only a subset of all potential regulators, namely the ones expressed in that particular tissue or cell line. The results page displays the number and proportion of query genes which are regulated by each trans-factor or regulatory element along with the enrichment p-values adjusted for multiple testing by the Benjamini-Hochberg method.

networks

By this mode, the user can interactively display and browse several known post-transcriptional networks, derived from the literature (experimental data only; no predictions included). This section will be constantly updated with new published networks. Users are encouraged to submit their own networks to be added to this part.

The user must first select a network from the dropdown box; the displayed network can then be zoomed, panned and rearranged by using mouse and keyboard controls (mouse wheel for zooming in/out, drag and drop for moving nodes, arrows keys for panning).

bioMart

This query mode allows to obtain all the data contained in AURA by exploiting the power of the BioMart query system. One needs to select a dataset to work on, one or more filter limiting the search space (e.g. the name of a trans-factor, or all data produced by a specific technique) and the desired output attributes.

The user may conduct search by UTR or Gene base datasets. By the UTR search the users obtains all UTR-related information, including identity of trans-factors, regulated UTRs and the positional information of the binding sites within the UTRs. By the Gene search the users obtains all Gene-related information, including uncoupling data and regulatory factors as well. One may construct the query by specifying various filter, such as, among the others, the type and identity of the regulatory element or the technique supporting the factor-UTR interactions that will be returned.

UTR view

In the “UTR view” all the information and annotations relative to a particular UTR are shown. When performing a search by target locus or trans-factor, the resulting UTR views are the same, with few minor differences.

The UTR view consists of a textual header presenting the annotations at the transcript level on top, followed by the “UTR browser”, displaying annotation at the UTR nucleotide level.

Uncoupling

This feature allows to see if the gene has been been involved in genome-wide studies profiling and comparing dynamic changes in the transcriptome and the translatome mRNA levels. Each row of the table corresponds to a different study.

Uncoupling is defined as the difference in the dynamic variation of a gene in the translatome with respect to the corresponding variation in the transcriptome. It is numerically measured as the difference between the translatome log2 fold change and the transcriptome log2 fold change of the gene in the experiment. An uncoupling value of 0 indicates that the gene shows exactly the same transcriptional and translational variations in the experiment. A positive uncoupling value denotes a gene that is translated more than what it would be expected looking at the variation in its transcriptome levels, while a negative value denotes a gene that is translated less than what it would be expected looking at the variation in its transcriptome levels. For a more detailed explanation of uncoupling, see also (PMID: 22672192).

Each uncoupling value is associated to a moderated t-test p-value, calculated with the Limma Bioconductor package, measuring for each gene the statistical significance of the difference between its variations in the translatome and in the transcriptome. P-values are corrected for multiple testing with the Benjamini-Hochberg method.

UTR view > The browser

The UTR browser is subdivided in two main parts:

• Textual header: it contains the chromosomal location, strand and length of the spliced-exons UTR, the name (HGNC) and description from UniProt/Swiss-Prot of the gene the UTR is belonging to. Other information is the “Overall conservation”, which is the mean PhastCons conservation score for the UTR, and the corresponding transcript half-lives. The supporting studies are indicated as tooltips on half-life numbers in order to allow the user to directly associate the half-life with the specific cell type or cell line of origin. A question mark icon denotes describing tooltips.
  
• UTR browser: it contains all the annotation related to the specific UTR. In particular, it contains multiple independent tracks that annotate the UTR by phylogenetic conservation, variation, trans and cis regulatory binding sites. The “Conservation” track shows the score calculated for each nucleotide in the 46 species alignment in the range 0 to 1. The “SNP” track shows the single nucleotide polymorphisms along the UTR sequence.Tracks relative to cis-regulatory sites are treated differently based on the type of search performed: if the query type is “trans-factor”, binding sites relative to those trans-factor are shown in a exclusive track, highlighted in magenta. The “RBPs” track contains RNA-binding proteins binding sites in orange and the “miRNA” track contains microRNAs sites in blue. Since for some of the trans-factors there is only partial information on the binding sites position we have provided dedicated track for such elements. A track whose title contains “unknown mRNA location” denotes trans-factor which binds the related transcript but without precise location; the trans-factor spans along the entire UTR with a thinner grey bar. A track whose title contains “unknown UTR location” denotes trans-factor which binds to a specific UTR type, but without precise location; the trans-factor spans along the entire UTR with a normal grey bar.

All annotations in the tracks are “clickable”: whenever the user clicks on an annotation, a description page containing all binding sites and related information is shown. By default, the UTR browser is zoomed on the full-length UTR. The user can then use the zoom controls to explore a smaller region within the UTR.

UTR view > Secondary structure

AURA provides a way to display the secondary structure of the selected UTR by clicking the corresponding button. The secondary structure is annotated as follow: base conservation scores are shown using a green scale gradient while SNPs locations are circled in red, RBP and microRNA binding sites are highlighted respectively in orange and blue. The magenta color is reserved for denoting binding sites of the query trans-factor (only in trans-factor search mode).

The interactive point and click approach and the nucleotide numbering along the UTR sequence allow the user to select nucleotides or nucleotide pairs of his choice as well as a variety of UTR structural annotations such as stems, helices and loops.

UTR view > Gene expression

This feature allows to explore gene expression levels related to the gene to which the selected UTR belongs, in more than fourty human tissues, as provided by GTEx (http://www.broadinstitute.org/gtex/).

The boxplots display the expression levels in the various tissues (one per x-axis value, as shown by the labels), quantified in terms of log10(RPKM). In case of trans factor search, whenever data is available, another plot containing gene expression levels for the trans factor is also shown.