The leucine rich repeat receptor like kinases (LRRKs) represent the largest subfamily of receptor like kinases (RLK), which are involved in diverse functions in plants (Shiu and Bleecker, 2001; Diévart and Clark, 2003; Shiu et al., 2004). Similar to the other RLK proteins, LRRKs also consisted of an extracellular domain (ECD) to perceive signals, a trans-membrane region to comfort the protein within the membrane, and a cytoplasmic protein kinase domain to stimulate the plant immunity via auto-phosphorylation, which is followed by phosphorylation of a specific substrate (Walker and Zhang, 1990; Shiu and Bleecker, 2001; Gou et al., 2010). The ECD of LRRKs consist of LRR domain, constituted by tandem repeats of 20–30 amino acid (AA) residues long leucine rich structural motifs (Jones and Jones, 1997; Matsushima and Miyashita, 2012). The LRR domain is stable among all proteins built from tandemly repeated motifs because of its hydrophobic solenoid inner core, where the conserved leucine and other aliphatic residues are present. This structure also favors the protein-protein interactions (Bella et al., 2008).

The LRRKs are involved in growth, development, and survival including organogenesis, morphogenesis, hormone signaling, and abiotic and biotic stress management in plants (Diévart and Clark, 2003; Li and Tax, 2013). For instance, BARELY ANY MERISTEM (BAM) 1–3 and CLAVATA1 are involved in apical growth, ERECTA-LIKE (ERL) 1–2 in organ growth, plant architecture, and stomatal and floral development, BRI1 in vascular development, HSL in flower abscission, FEI2 in cell wall biosynthesis, and GSO and RPK1 in embryogenesis, (Clark et al., 1997; Delgado et al., 2004; Shpak et al., 2004; Deyoung et al., 2006; Nodine et al., 2007; Stenvik et al., 2008; Tsuwamoto et al., 2008; Xu et al., 2008). The LRRK FLS2 and EFR regulate the plant immunity and act as receptor for flagellin and EF-Tu for plant defense (Gomez and Boller, 2000; Zipfel et al., 2006). PSKR1, BAK1, BRI1, and SERK of arabidopsis are also reported to be involved in biotic stress and defense responses (Nam and Li, 2002; Roux et al., 2011; Wang et al., 2011; Mosher et al., 2013; Wu et al., 2016). The differential expression of LRRK genes during abiotic stresses in various plants like arabidopsis, rice, and potato has been reported (Diévart and Clark, 2004; Osakabe et al., 2005; Park et al., 2014; Wu et al., 2016). The GHR1 of arabidopsis is induced by ABA and H2O2, while RPK1 by drought, salt, abscisic acid (ABA), and low temperature (Osakabe et al., 2005; Hua et al., 2012). OsGIRL1 is involved in salt, osmotic, heat, and gamma radiation stress in Oriza sativa (Park et al., 2014). Several other differentially expressed LRRKs are known in response to various biotic and abiotic stresses in plants, whose specific functions need to be established in future studies. Moreover, some LRRKs like SOMATIC EMBRYOGENESIS RECEPTOR-LIKE KINASE (SERK) and ERECTA of arabidopsis (Godiard et al., 2003; Shpak et al., 2004; Colcombet et al., 2005) possess dual functions, which are due to either recognition of numerous ligands by a single receptor or cross talk between various pathways (Torii et al., 1996; Afzal et al., 2008). The LRRKs are mostly evolved by extensive gene/domain duplication; therefore they have probably acquired diverse functions through neofunctionalization (Jones and Jones, 1997; Walsh, 2003).

In plants, the functional characterization of LRRK genes are largely performed in model plant arabidopsis (Shiu and Bleecker, 2001; Shiu et al., 2004; Wu et al., 2016). In recent years, the study has been certainly extended, but it is still limited to a few plants like rice, tomato brassica, poplar, and soybean (Sun and Wang, 2011; Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015; Zhou et al., 2016). It is probably due to lack of genomics information, large number of genes and complication in applicability of functional genomics tools in various other plants. The Triticum aestivum (bread wheat) is one of the main food crops of the world, which is being consumed by about one-third population as a major source of calorie. It is used in the preparation of bread, biscuit, cake, pastries, and several other food materials. The T. aestivum comprises a composite allohexaploid genome (2n = 6x = 42; AABBDD), which was originated by hybridizations of three diploid A, B, and D sub-genomes (Marcussen et al., 2014). The characterization of LRRK gene family has not been performed in T. aestivum, despite being a staple food crop. The International Wheat Genome Sequencing Consortium (IWGSC) has reported the chromosome based genome information of T. aestivum in 2014 (IWGSC, 2014). Further, numerous developmental stage specific and stress related high-throughput RNA sequence (RNA seq) data of T. aestivum have been reported in recent years (Zhang et al., 2014, 2016; Liu et al., 2015; Pingault et al., 2015). The availability of genomic information and these RNA seq data enabled us for the identification and characterization of LRRK gene family in T. aestivum. Since the LRRK genes play vital roles in development and stress tolerance in plants (Shiu et al., 2004), their characterization will be valuable in future genetic improvement program of T. aestivum.

Herein, the entire TaLRRK gene family was identified using the genomic data of T. aestivum (IWGSC, 2014) and comprehensive analyses were performed. The gene family was characterized for genome wide distribution, occurrence of duplication events, homologous, and orthologous genes identification, annotation, and gene ontology (GO) mapping, gene, and protein structure, domain, and motif organization, and evolutionary relationships. The expression profiling during various developmental stages and stress conditions was performed to understand their role in these biological processes. To the best of our knowledge, this is the first report on genome wide characterization of LRRK genes in T. aestivum.

The LRRKs comprise a major subfamily of RLK. The LRRK genes play vital role in majority of biological activities of plants including growth and developments to stress management (Shiu and Bleecker, 2001; Shiu et al., 2004; Diévart and Clark, 2003). However, their functional characterization has still not been performed in T. aestivum, the most important crop plant of the world. This could be due to several factors like complex allohexaploid (2n = 6x = 42; AABBDD) genome, unavailability of genome sequence data and others. The availability of chromosome based genome information, and various developmental stage related and stress responsive RNA seq data in recent years facilitated the genome wide characterization of various gene families in T. aestivum (IWGSC, 2014; Zhang et al., 2014, 2016; Liu et al., 2015; Pingault et al., 2015; Shumayla et al., 2016). Since the LRRKs are responsible for various essential functions in plants (Shiu et al., 2004), here we have carried out a comprehensive analysis of these genes in T. aestivum.

Blast search of known protein sequences from model plants like arabidopsis, HMM blast and pfam blast are the standard procedures for the identification of new gene family sequences from non-model plants. Similar procedure was followed to characterize LRRK genes from various plants earlier (Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015). Here, we identified 531 non-redundant TaLRRK genes following the similar procedure, which were further confirmed by blast search against NCBI-conserved domain, PROSITE-Scan and SMART databases. Since the LRRK is the largest subfamily of RLK (Fischer et al., 2016), the number of TaLRRK genes identified was consistent with our expectation. About 213, 234, 303, 309, 382, and 467 LRRK genes are earlier reported in arabidopsis, tomato, rice, brassica, poplar, and soybean genomes, respectively (Shiu and Bleecker, 2001; Shiu et al., 2004; Sun and Wang, 2011; Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015; Zhou et al., 2016). The number of TaLRRK genes was fairly higher than the earlier studied plants, which might be due to the large hexaploid genome of T. aestivum (IWGSC, 2014). Similarly the higher number (467) of LRRK genes are reported from soybean, another polyploid crop plant (Schmutz et al., 2010; Zhou et al., 2016).

The allohexaploid genome of T. aestivum was originated as a result of hybridization of three (A, B, and D) diploid sub-genomes (Marcussen et al., 2014), which indicates the contribution of each sub-genome in composition of various gene families of T. aestivum (Shumayla et al., 2016; Zeng et al., 2016). Further, the contribution of number of genes from a sub-genome would be similar to their number in progenitor genome, unless they are not duplicated or deleted in due course of evolution after polyploidization. The genome wide distribution analysis indicated that the TaLRRK genes were also derived from individual sub-genomes (Figure 1A), while their number on A and D sub-genomes was quite lower than the number of LRRK genes predicted in T. urartu and Aegilops tauschii, which are progenitors of these sub-genomes, respectively (Marcussen et al., 2014). This indicated deletion of certain TaLRRK genes from these sub-genomes during evolution of T. aestivum after hybridization. The TaLRRK genes were also found to be distributed on each chromosome but at diverse frequency, as earlier reported in various other plants (Rameneni et al., 2015; Wei et al., 2015). For example, maximum and minimum number of TaLRRK genes was located on chromosome 2B and chromosome 5A, respectively. Similarly, chromosome 5 and chromosome 4 in arabidopsis, and chromosome 1 and chromosome 10 in brassica contain the highest and lowest number of LRRK genes, respectively (Rameneni et al., 2015; Wu et al., 2016).

The probability of occurrence of homologous genes has increased thrice in T. aestivum due to its allohexaploid genome (AABBDD). Hypothetically, each T. aestivum gene should have a minimum of three homologous genes, at least one from each sub-genome. These may also called as homeologous genes because of their localization on homeologous chromosomes. Here, we observed 531 TaLRRK genes in T. aestivum, which could be grouped into 233 distinct clusters of one or more homologous genes. The total number of TaLRRK genes was not actually thrice to their number in progenitor genomes like T. urartu and Ae. tauschii. Further, we could not detect three homeologous genes in case of each cluster. The majority of TaLRRK genes were single without any homeologous/homologous sequences, while a few had large number (up to 20) of homologous genes. This might be due to the consequences of deletion and/or duplication of certain TaLRRK genes during the process of evolution and acclimatization of T. aestivum. Similar kinds of deletion and duplication events are also reported in other gene families T. aestivum (Shumayla et al., 2016; Zeng et al., 2016). Moreover, the number of distinct TaLRRK genes clusters was comparable to the number of LRRK genes reported\predicted in various other diploid plants like arabidopsis, rice, T. urartu, Ae. tauschii and others (Shiu et al., 2004; Sun and Wang, 2011; Rameneni et al., 2015; Wei et al., 2015).

Genome hybridization, duplications, and inter-chromosomal rearrangements of various genes are reported as major events during the evolution of T. aestivum genome (Choulet et al., 2010; Marcussen et al., 2014; Akpinar et al., 2015; Glover et al., 2015). The role of paralogous genes evolved by duplication events (DE) are also reported in the evolution of lectin receptor kinases of T. aestivum (Shumayla et al., 2016), which indicated the role of duplication in the evolvement of other sub-families of RLK in this plant. Therefore, the frequency of various DEs was also analyzed in case of TaLRRK genes, following the established method (Shumayla et al., 2016). A large number of DEs were predicted in case of TaLRRK genes, which were distributed on various sub-genomes and chromosomes. About 25% of DEs were classified as TD and SD, while the majority of them could not further categorized due to the localization of one of both paralogous gene(s) on scaffolds. This is probably due to the unavailability of complete genomic data of T. aestivum. The occurrence of repeat sequences and transposable elements in the genome positively favors the gene duplication during evolution (Magadum et al., 2013). The T. aestivum genome is highly rich in these elements (Choulet et al., 2010, 2014; Marcussen et al., 2014; Glover et al., 2015), which might encourage the DEs in various gene families including TaLRRKs. The role of DEs in the expansion of LRRK gene family is earlier reported in various other plants like poplar, tomato, brassica and arabidopsis (Shiu and Bleecker, 2001; Shiu et al., 2004; Sun and Wang, 2011; Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015). These results along with earlier reports from other plants indicated that the gene duplication played critical role in diversification and neofunctionalization of LRRK genes during evolution.

The orthologous genes of TaLRRKs in rice and arabidopsis were identified by bidirectional best hit approach. Since both monocots and eudicots are evolved from common ancestor (McClung, 2010), the evolutionary relationship in the form of orthologous genes is obvious between them. Such kind of relationship has been reported in various gene families and plants in earlier studies (Sun and Wang, 2011; Rameneni et al., 2015; Wei et al., 2015; Shumayla et al., 2016).

The functional annotation of TaLRRK genes by blast search against various databases like NCBI-nr, UNIPROT and UNIREF further validated their identity. The TaLRRK sequences showed homology with most of the functionally characterized LRRKs like BRI1, ERL, SERK, CLAVATA from various plants including model plant arabidopsis. These LRRK proteins are earlier reported to be involved in various developmental and stress related activities (Wu et al., 2016). The GO mapping provides information about the characteristic features of various genes across the taxa. Since the LRRKs are membrane proteins (Torii, 2004), the GO analysis indicated membrane bound nature of each TaLRRK protein. Further, the prediction of binding and catalytic properties in TaLRRK proteins supported ligand interacting and enzymatic nature of LRR and kinase domains, respectively (Diévart and Clark, 2004; Torii, 2004). The GO mapping also predicted TaLRRKs role in signaling, stimulus response, and various other metabolic processes, which depicted their diverse function in plant development and defense mechanism. Similar GO categorization is also reported for brassica LRRKs (Rameneni et al., 2015).

The gene structure analysis showed variation in the presence of number of introns in various TaLRRK genes. The number of introns in TaLRRK genes was varied from zero to a maximum of 27 introns, wherein the majority of genes were single intronic. The variation in the number of exon/intron composition in LRRK genes is reported earlier in various plants. The single intronic LRRK genes are also reported as abundant class in rice, tomato, and poplar. Further, a maximum of 26 introns were reported in poplar, rice and tomato, while 24 in brassica (Sun and Wang, 2011; Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015). Similar complexity in ERECTA gene structure was reported in arabidopsis (Karve et al., 2011), which was predicted as an ortholog of LRRK69.1 gene of T. aestivum. Previous studies suggested that introns influence the expression of genes, which has been experimentally validated with ERECTA gene in arabidopsis (Morello and Breviario, 2008; Karve et al., 2011; Chorev and Carmel, 2012). The significant expression of a few high intronic TaLRRK genes was also observed in certain tissues; however we could not establish a correlation between number of introns and expression pattern.

The length, molecular weight, and pI of TaLRRK proteins also varied as reported in various other plants (Sun and Wang, 2011; Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015). Since the majority of RLK are membrane protein and associated with the plasma membrane (Torii, 2004). Therefore, presence of a transmembrane region and signal peptide is the signature of the membrane RLK (Shiu and Bleecker, 2001). Occurrence of one or more transmembrane regions and a signal peptide indicated membrane bound nature of most of the TaLRRK proteins and majority of them were localized on plasma membrane. This indicated similar nature of TaLRRK proteins as reported in other plant species (Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015).

Analysis of domain composition showed the occurrence of 12 different types of domains in TaLRRK proteins, which were found in various combinations, where as kinase domain was consistently present at the C-terminus of each protein (Figure 3B). These diverse domain compositions also indicated functional divergence in various TaLRRK proteins. The diversity in domain composition has also been reported in other plant species like rice and brassica (Sun and Wang, 2011; Rameneni et al., 2015). Mostly the variation in domain composition is found at N-terminus and especially in extra-cellular domain of LRRKs, which supported the deviation in types of signal perception by LRRK proteins. This might be a possible reason for their diverse biological role in plant life cycle. Most of the TaLRRK proteins consisted of numerous types of LRR domains. Though the specific function of each type of LRR domain is not known, but largely they are tandemly repeated leucine rich motifs involved in protein-protein interactions (Bella et al., 2008). Malectin and P-loop NTPase domains were also found in a few TaLRRK proteins, which are probably involved in interaction with glycan and nucleic acid moieties (Leipe et al., 2004; Schallus et al., 2008), however their exact role in LRRK proteins is not established. Similarly a few TaLRRKs consisted of Aldolase II and Herpes_gE domains with unknown function. Several conserved motifs were also predicted in both extracellular LRR and intracellular kinase domains (Figure 3C; Table S2). Most of these motifs are also present in various other plant species (Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015). The organization of conserved amino acids (GxIPxxLxxLxxLxxLxLxxN) in LRR motif was also found to be highly conserved in both monocot and dicot plants like rice, tomato, and others (Sun and Wang, 2011; Rameneni et al., 2015; Wei et al., 2015). A few conserved motifs were found near the important sites like ATP binding sites, substrate binding sites, and activation loop of kinase domain, which play critical role during signal transduction (Clouse, 2011).

The phylogenetic relationship was analyzed using full-length TaLRRK protein sequences, because it provided excellent evolutionary inference. Full length sequences were used in case of LRRK phylogenetic analyses of other plant species (Sun and Wang, 2011; Rameneni et al., 2015; Wei et al., 2015). The identified homeologous and/or homologous sequences were closely grouped, which further established high sequence homology and evolutionary relation between them. The predicted paralogous sequences were also clustered together, which indicated their similar origin. Similar domain architecture and LRR motifs organization in most of the clades indicated functional correlation between these proteins, which has been observed during expression analysis and discussed below. Further, comparable organization of exon/intron pattern in various sub-groups and clades indicated a strong evolutionary association between them at gene level also. Similar type of gene structure and evolutionary relationship between LRRKs are also reported in other plant species like tomato, rice brassica and poplar (Sun and Wang, 2011; Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015).

The expression of a gene in a particular organ can be used as a source of information for identifying its function in that organ. Thus, the expression data of 531 TaLRRK genes was extracted from the available whole transcriptome dataset of T. aestivum from five different organs (root, stem, leaf, grain, and spike), each with three developmental stages (Pingault et al., 2015). The genes with at least 10RPKM expression values in one or more developmental stages were used for detailed analysis. It might be possible that these genes played critical roles in development of various organs, while the remaining genes might be involved in other functions like stress management. The organ specific expression indicated divergent role of various TaLRRK genes. Most of the genes were specifically expressed in particular organ like leaves or roots or grains, which indicated their specific role in the development of that organ. The organ specific variation in expression of LRRK genes is also reported in arabidopsis, tomato, brassica, soybean, and poplar (Gou et al., 2010; Zan et al., 2013; Rameneni et al., 2015; Wei et al., 2015; Zhou et al., 2016). An evolutionary functional association between organ specific TaLRRK genes was also established as majority of high expressing genes from a particular organ were clustered into same phylogenetic group.

The developmental stage specific expression profiling in various organs indicated that the genes, which were earlier categorized as organ specific, were found to be developmental stage specific. Since the functions of various LRRK proteins of arabidopsis have been established (Wu et al., 2016), the putative functions of few TaLRRK proteins were predicted based on their sequence homology. The arabidopsis SERK2 reported to be involved in anther development, maturation and brassinosteroid signaling in roots, anthers, and seedlings (Colcombet et al., 2005; Roux et al., 2011; Wu et al., 2016), showed homology with LRRK94.3, which was also highly expressed in root, grain and spike. Similarly, various other proteins of arabidopsis showed sequence homology with TaLRRK proteins such as BAM1-3 (LRRK60, 122, and 177) and CLAVATA (LRRK41) that are involved in shoot and floral growth, and cell differentiation, GSO (LRRK101) and RPK1 (LRRK116.2) in embryo development, ERL1-2 (LRRK69.2) in organ growth and floral development, BRI1 (LRRK90.1 and 90.4) in shoot vascular development, HSL2 (LRRK107) in abscission of flower and FEI2 (LRRK192.1) in cell wall biosynthesis (Clark et al., 1997; Delgado et al., 2004; Shpak et al., 2004; Deyoung et al., 2006; Nodine et al., 2007; Stenvik et al., 2008; Tsuwamoto et al., 2008; Xu et al., 2008).

The expression pattern of TaLRRK genes was also studied using high-throughput transcriptome data generated after 24 h of infection of two common fungal pathogens Bgt and Pst, which are responsible for powdery mildew and stripe rust diseases, respectively (Zhang et al., 2014). We observed a differential response of TaLRRK genes for these two pathogens though both of them are fungal species. This indicated that a diverse mechanism of signal perception and transduction is involved in these two pathogen infections. Further, several highly affected genes during fungal infections were actually very low abundant in normal plant developmental stages, which established their specific role in these biotic stress responses. The differential expression patterns of LRRK genes under fungal attack have also been reported in other plants like poplar (Zan et al., 2013). Though none of the earlier studies have reported characterization of LRRK genes during Pst and Bgt infection, several arabidopsis LRRK proteins are reported to be involved in biotic stress management (Wu et al., 2016). LRRK proteins like PSKR1, BAK1, BRI1, SERK, and FLS of arabidopsis that are involved in various defense responses (Gomez and Boller, 2000; Nam and Li, 2002; Roux et al., 2011; Wang et al., 2011; Mosher et al., 2013; Wu et al., 2016), showed homology with LRRK126.1, 90.3, 188, 39.3, and 220 of T. aestivum, respectively. This indicated that these genes might have similar function as their arabidopsis homologs, which needs to be functionally characterized in future studies.

Plants are exposed to various abiotic stresses like heat, drought, salt and others, which drastically affect the yield by distressing various biochemical and physiological processes (Prasad et al., 2011; Pradhan et al., 2012; Izadi et al., 2014; Yang et al., 2014). Since the LRRKs are reported to be involved in abiotic stress responses (Torii, 2004), we analyzed the effect of various abiotic stresses on the expression of TaLRRK genes using high throughput transcriptome data (Liu et al., 2015; Zhang et al., 2016). Since heat and drought generally act in synergistic manner (Liu et al., 2015), the expression of TaLRRK genes was analyzed under heat, drought, and their combined stresses as well. Several TaLRRK genes were found to be differentially expressed, which indicated their role in these stress conditions. Diverse response of TaLRRK genes during DS and HS indicated variation in their modus operandi. However, a similar trend of up or down regulation was observed for most of the affected genes during HS and DS. In case of HD, majority of genes showed assorted expression, which revealed cross talk between signaling pathway during these stresses. Majority of up-regulated TaLRRK genes were very low expression during normal condition, which indicated their role in these stresses. The phylogenetic analysis revealed that most of the highly affected genes were clustered into group VII (Figure 6D). It is noteworthy that the group VII consisted of least expressing genes during normal stages of various organ development (Figures 5C-G). The results indicated that these group VII genes were not only associated in their expression pattern, but also evolutionarily linked with each other. The role of LRRK genes in abiotic stress management is reported in various plants like rice and potato (Diévart and Clark, 2004; Park et al., 2014; Wu et al., 2016). The OsGIRL1 of rice, which showed homology with LRRK29.1 of T. aestivum, was earlier reported to be involved in heat, salt, osmotic, and gamma radiation stresses (Park et al., 2014).

Salt stress adversely affects the productivity of several crops, and it has been earlier reported that the amount of land affected by salinity is increasing (Wang et al., 2003). Salt stress induces various complex biochemical, molecular, cellular, and physiological changes in plants (Wang et al., 2003; Tuteja, 2007; Munns and Tester, 2008). Salt induced changes in transcriptome expression pattern are earlier reported in the root of T. aestivum (Zhang et al., 2016). The TaLRRK genes were also found to be affected by salt stress. Interestingly, the affected TaLRRK genes could classify into two groups with constant up and down regulation with the exposure time of salt. Most of these up and down regulated genes were clustered into phylogenetic group VII and III, respectively. The results indicated evolutionary association between the genes showing similar behavior during salt stress. The role of LRRK genes in salt stress management is also reported in rice and arabidopsis (Ten Hove et al., 2011; Park et al., 2014).

In conclusion, the present study was focused on the analyses of various characteristic features and expression patterns of TaLRRK genes of T. aestivum, a major food crop. Similar to the other plant species, TaLRRK was found to be one of the largest gene sub-family, which was categorized into eight distinct phylogenetic groups. The gene and protein structure, domain architecture, motifs arrangement, and composition were mostly conserved in various phylogenetic sub-groups. Occurrence of numerous paralogous genes indicated the role of duplication events in the expansion of this gene family during the evolution process. The expression profiling indicated the role of TaLRRK genes in development of various organs under normal and different stress conditions, however the specific role of individual gene needs to be validated by functional genomics tools in future studies. The analysis of major responsive genes in various stress conditions indicated specific roles of most of the TaLRRK genes. For instance, 33, 35, and 31 out of top 50 affected TaLRRK genes were specifically related to biotic, salt, and heat/drought stress, respectively (Figure 6E). Moreover, LRRK134.1, 134.2, and 134.3 genes were affected during all stress conditions. These results may provide valuable information to investigate the function of individual TaLRRK genes in plant development and stress responses using modern genome editing and functional genomics tools.

SU conceived the idea and designed the experiments. S, SS, and RK performed the experiments. S, VM, KS, and SU analyzed the data. S and SU wrote the manuscript. VM and SU finalized the manuscript.