To integrate previously published plant EST data, all Putative Unique Transcripts (PUT) representing 273 transcript assemblies were downloaded from PlantGDB (Version release 187) available from http://www.plantgdb.org/ (Dong et al., 2004). Additionally, version control 74 NGS transcriptomes available at PhytoMetaSync² (Facchini et al., 2012; Xiao et al., 2013), 14 medicinal plant transcriptomes available from medicinal plant genomics resource (MPGR)³ (Góngora-Castillo et al., 2012; Góngora-Castillo and Buell, 2013) and 3 Brachypodium sylvaticum transcriptomes available from http://jaiswallab.cgrb.oregonstate.edu/genomics (Fox et al., 2013) were downloaded, representing a total of 364 plant species.

For systematic identification of SSR, all the transcripts (ESTs as well as NGS) assemblies were first scanned for the presence of the homopolymer errors and sequence ambiguity was removed using the est_trimmer tool available at: http://pgrc.ipk-gatersleben.de/misa/download/est_trimmer.pl with the following settings: -amb = 2.50 -tr5 = T, 5.50 -tr3 = A, 5.50. Following the transcript ambiguity removal and trimming of the homopolymer runs, MISA (MIcroSAtellite identification tool) (Thiel et al., 2003) was deployed to identify the microsatellites. In the present version of the PlantFuncSSRs, we classified microsatellites as repetitive stretches of motifs of a minimum and 12-mer repetitive stretch as mono-, 6-mer repetitive stretches as di, 4-mer repetitive stretches of tri- and tetra-, and a minimum of 3-mer repetitive stretch as penta- and hexa-nucleotide. Additionally, the identified SSRs have been classified into perfect and compound repeats, with compound repeats interrupted by a minimum of 100 bp as previously described (Victoria et al., 2011). Primer pairs were designed for all of the identified SSRs using primer3 available from primer3.sourceforge.net (Untergasser et al., 2012) using the settings as described in MISA (Thiel et al., 2003).

Following SSRs identification, in-depth functional annotation of the identified SSRs was carried out using the standalone annotator Sma3s (Muñoz-Mérida et al., 2014), which uses the plant taxonomic division set in the Uniprot database⁴, including both Swiss-Prot and TrEMBL sections to enrich the final annotation. The annotations gave the found Gene Ontology (GO) terms which were subsequently linked to their GO_SLIM terms using the plant GO slim available from www.geneontology.org, in order to simplify the GO terms and allow cross-comparison. In this way, each SSRs sequence was identified with the more probable gene name and description, as well as both GO terms from the existing three categories and Swiss-Prot keywords, all of them for cataloging the SSRs and assigning functional domains. The IntAct annotations and Interactions were cross-linked using the IntAct resources available from EBI at: http://www.ebi.ac.uk/intact/. The functional SSRs annotation also includes putative InterPro domains (Quevillon et al., 2005, pathways from UniProt to have more details of the involved biological processes. PlantFuncSSRs presents only those SSRs, which have functional annotations appended to them and are thus termed as SSR-functional markers.

Expressed sequence tags and NGS based Transcriptome reconstruction represent the functional portion of the genome and have been widely used as resources to mine and develop functional markers. Developing an efficient browsing system for the mining of repeats is an important task, as this can be widely applied to a wide range of on-going plant breeding and crop improvement research. To develop an efficient browsing system, PlantFuncSSRs architecture has been developed using Ruby Rails and MySQL, which provides faster integration and query based searches to the users. The current version of the PlantFuncSSRs presents more than 2 million SSRs and SSR-FDMs from 364 species for easy access and browsing of transcript derived plant SSRs across the plant kingdom (Table 1). These species are ranging from important crops to wild species, from mono- to di-cots, from annual to polyannual and wood species. Integration of visualization features with the rapid mining of the data is a key central feature that has been implemented in the PlantFuncSSRs. A schema of the database architecture in the form of entity-relationship is given in Figure 1. For the visualization of the SSRs and the associated information, several hierarchal levels of classified information have been inter-linked in PlantFuncSSRs (Figure 2). The front-end portal is user-friendly and allows the end-users to search SSRs as “species-wise”, “family wise”, or “advanced search menu” (Figure 2). A quick search implementation pattern displays the embedded species information in quick select “species” and “families”, which are hyperlinked pages to the respective species and provide a quick view of the functional SSRs present in each species. Figure 3 shows the webpage browsing of PlantFuncSSRs with detailed classification of the identified SSRs for user-selected species of interest. Alphabetical classification of the species provides an additional advantage for the users to quickly look for their species of interest (Figure 3).

Each record in the species displays the Species_Name, Num_Seqs_Exam, Size_Exam_Seqs, Num_SSR_Ident, SSR_Cont_Seqs, Seqs_Cont_SSR, and Num_SSR_Present providing summarized information on the number of the identified SSRs for that particular species of interest lined to the primer pair information and high throughput functional annotation (Figure 3). In PlantFuncSSRs, each species page has been hyperlinked to the corresponding repeat information pages that present detailed information on several statistics such as total number of sequences examined, total size of examined sequences (bp), total number of identified SSRs, number of SSR containing sequences, number of sequences containing more than one SSR and compound SSRs (Figure 3). In addition, to this summary information, each species classified page also details the types and distribution of the repeats in tabular format, which can be sorted “on the fly”. An integral part of PlantFuncSSRs is to describe the associated primer pair information for each species to facilitate the development of functional SSRs for diversity analysis. To augment such capacity, each functional SSR has been associated with primer pages and detailed functional annotations, which describes the set of the “ready to use” primers for the functional validation of the corresponding SSRs (Figure 4).

Microsatellites (SSRs) have been shown to be regulators of a number of plant genes demonstrating their importance as key players in regulating plant function (Faville et al., 2004). FuncPlantSSRs offers a wide variety of functional annotations for the identified SSRs such as GO terms, GO slim categories, pathways, descriptions to identify the sequences and comparing with putative homologues, and motif and domain modules to offer the domain architecture for the sequences. Recently, increasing interest toward the functional linkage of the markers to the domain association and function can be seen from several recent reports in plants such as Ocimum basilicum (Gupta et al., 2010), Seasmum indicum (Bhattacharyya et al., 2014), Elaeis guineensis (Tranbarger et al., 2012), and Camellia sinensis (Sahu et al., 2012) suggesting the role of the functional SSRs as important markers for developing the functional genic approaches for marker enrichment in plants. Nonetheless, established reports of the functional association of the repeats with the catalytic domains (Parida et al., 2010; Yu et al., 2010) has been widely developed. For quick advanced searches, PlantFuncSSRs offer several functionalities, such as searches customized and optimized on various hierarchal levels i.e., Family, Species, Type of Repeat, Number of Repeat, Functional annotation, GO annotation, and IPR annotations (Figure 2). Availability of the curated information provides end users with the flexibility to narrow their searches to functional SSRs linked to specific categories, motif types or functional annotations. Taking into account the vast amount of the species coverage and associated functional SSRs present in the PlantFuncSSRs, we believe that the PlantFuncSSRs provides access to the most comprehensive catalog available for the functional SSRs from plant transcriptomes.