Valid data input examples
Transcripts
When using the transcript query option, the database returns results for transcripts as potential targets of candidate miRNAs. Valid entries consist of locus identities as defined by TAIR.
Examples of valid transcript IDs are:

AT1G01010.1
at1g01010
Intergenic regions
To perform an intergenic query, enter the ID for the gene immediately South of the intergenic region on the chromosome. When using the intergenic query option, the database returns candidate precursors predicted to be present within the intergenic region. These results will contain potential targets encoded from many regions of the genome.

(N.B. Some browsers (e.g. Camino ) may not be able to display all the results for a large intergenic region)

Introns
Examples of miRNAs resulting from the processing of introns are known for animal but not yet for plants. Despite this, the database will accept intron queries. Valid intron IDs follow those used by TAIR

An example of an Intron query is:

AT3G44280.1-I2

For intron queries, only results for the transcript strand are returned.
Sequences
The database can also be queried using a nucleotide sequence. Records for which the precursor candidate sequence completely contains the query sequence will be returned. The minimum allowed length of the sequence query is 15 nucleotides.

Examples include:

ugacagaagagagugagcac

UGACAGAAGAGAGUGAGCAC

GTCTAGATTAGGGTTTATGGCAATAGAGCTGCTTTGTTCGTGATTTTAGGAATTGGGTATTGAG



Show All

This option will return all predicted hits without filtering. Warning - This can result in the return of more than one hundred hits.






5' U is Present

 With this box checked the results will contain only those records for which a U residue is present in the -2, -1, +1 and +2 positions with respect to the 5' nucleotide of the predicted miRNA sequence. The range of positions is provided to allow for the inability of findMiRNA to accurately predict the 5' end of the miRNA sequence based on miRNA-target base pairing.






Target-miRNA Minimum Free Energy threshold

 This sets the lowest stability limit for the predicted miRNA-target pair as estimated from the minimum free energy of the miRNA::target duplex using Mfold 3.1. The default value will select records containing miRNA-target MFE values that are equal or more negative than -28 kcal per mol.






Result Limit

 This parameter allows an arbitrary cut off for the ranking number of the returned candidates. The ranking number describes the initial ranking of the candidates targeting a single transcript before structural analysis with RNAFold. The candidate ranking depends on 2 things: how well the candidate targets the mRNA, and the binding of the miRNA in the hairpin. In general, known miRNAs have a ranking number < 125 and will be passed by a Result Limit of 125. The maximum Result Limit is 500.






Has predicted homologues in rice

 Homologues were predicted using precExtract, a supplemental program that aligns predicted/hypothetical miRNAs to a genome. The PrecExtract alignments included in this dataset were aligned to Rice with 0 mismatches. In general, known miRNAs have an ortholog in Rice.






GC content of miRNA

 Allows filtering based on the percent GC of the predicted miRNA sequence.






Overlapping target sites

 When checked, this option will result in the return of records for which the predicted miRNA target sites overlap. This option allows for a difference of plus or minus 2 nucleotides for the target site overlap and will enable screening for potential miRNA precursor families that should have the same target site within a particular transcript. Records with overlapping target sites are also indicated by green values in the 'Transcript Bounds' field. However, since this mark up is done before any filtering is performed, some of these records may later appear on their own when neighboring records have been removed after filtering is applied.






miRNA and sequence query overlap

 This option is for use with sequence queries (DNA/RNA sequences). When checked, the results returned will only be those for which the miRNA and query sequence overlap by at least 15 nt.






Normalized Precursor MFE

 This option allows the user to specify the length normalized precursor minimumfree energy as determined from the Vienna RNAFold package. The default values remove 75% of the data, but retains all of the previously reported miRNA precursors.