161
orders used multiple times where required. Among the three families of siderophores, fusarinines and coprogens generally contain only a single module whereas
ferrichromes might contain three or more of such modules. In any case, typically the
final module or the only module present is found to be consisting of additional ‘T’
or ‘C’ modules (Schwecke et al. 2006). In addition, it is found that in the case of
siderophore BGCs, genes homologous to ABC transporters – expected to be
involved in secretion functions of siderophores – are often found clustered with core
genes (Johnson et al. 2013; Munawar et al. 2013). Thus taking clues from these
ideas, fungal genomes could be mined for potential siderophore synthesis genes.
Even though siderophore-specific prediction tools for fungi are not yet developed, identification of intact NRPS coding genes/individual NRPS domain from
genomic data was first put forward as early as in 2004 from India via the interface
NRPS-PKS (Ansari et al. 2004). The first version of the interface was developed by
studying 22 BGCs that were experimentally characterized. The NRPS domains
were manually curated and used to create a domain database. To identify similar
domains in the input sequence, local alignments to the database using BLAST were
conducted. In the past 15 years, the techniques for annotation have improved significantly. The implementation has since been updated and currently executed as a
server called SBSPKSv2 that provides a platform for comprehensive analysis of
secondary metabolite BGCs (Khater et al. 2017). With more than 130 characterized
BGCs, HMM models for better searches and the inclusion of structural analysis, the
server is a user-friendly interface for detection of putative NRPS/PKS BGCs and
also provides clues on the putative products of uncharacterized NRPS/PKS clusters.
Such a platform thus facilitates the identification and annotation of putative siderophore genes from fungal genomes.
Nevertheless, there are other popular bioinformatics tools developed afterwards
based on similar principles but using better datasets in terms of number, specificity
and diversity. The tools antiSMASH (Medema et al. 2011; Blin et al. 2017) and
SMURF (Khaldi et al. 2010) are hence set off as the current state-of-the-art tools for
BGC identification. The former is tailored to incorporate both bacterial and fungal
genomes whereas the latter is specific to fungal genomes only. However, SMURF is
only capable of gene prediction and hence must be used with other tools for the
purpose of domain localization. Though these tools have been commonly used for
prediction of BGCs producing various NPs and secondary metabolites in fungi, they
have not been well exploited for siderophores.
As important as the localization of the BGC gene is the prediction of their substrate, particularly to isolate siderophore-producing genes. To facilitate the studies
of “cryptic” NRPS clusters that are not linked to any NP, prediction of substrate
specificity could be conducted (Verne Lee et al. 2015). This could be achieved by
analysis of the substrate-binding adenylation domain ‘A’ in the NRPS module.
Since domain architecture of siderophore-NRPS genes are not unified by a common
pattern, database search approaches have been found useful (Etchegaray et al. 2004;
Komaki et al. 2018; Aleti et al. 2015). Databases are developed based on sequence
10 Bioinformatics Applications in Fungal Siderophores: Omics Implications
orders used multiple times where required. Among the three families of siderophores, fusarinines and coprogens generally contain only a single module whereas
ferrichromes might contain three or more of such modules. In any case, typically the
final module or the only module present is found to be consisting of additional ‘T’
or ‘C’ modules (Schwecke et al. 2006). In addition, it is found that in the case of
siderophore BGCs, genes homologous to ABC transporters – expected to be
involved in secretion functions of siderophores – are often found clustered with core
genes (Johnson et al. 2013; Munawar et al. 2013). Thus taking clues from these
ideas, fungal genomes could be mined for potential siderophore synthesis genes.
Even though siderophore-specific prediction tools for fungi are not yet developed, identification of intact NRPS coding genes/individual NRPS domain from
genomic data was first put forward as early as in 2004 from India via the interface
NRPS-PKS (Ansari et al. 2004). The first version of the interface was developed by
studying 22 BGCs that were experimentally characterized. The NRPS domains
were manually curated and used to create a domain database. To identify similar
domains in the input sequence, local alignments to the database using BLAST were
conducted. In the past 15 years, the techniques for annotation have improved significantly. The implementation has since been updated and currently executed as a
server called SBSPKSv2 that provides a platform for comprehensive analysis of
secondary metabolite BGCs (Khater et al. 2017). With more than 130 characterized
BGCs, HMM models for better searches and the inclusion of structural analysis, the
server is a user-friendly interface for detection of putative NRPS/PKS BGCs and
also provides clues on the putative products of uncharacterized NRPS/PKS clusters.
Such a platform thus facilitates the identification and annotation of putative siderophore genes from fungal genomes.
Nevertheless, there are other popular bioinformatics tools developed afterwards
based on similar principles but using better datasets in terms of number, specificity
and diversity. The tools antiSMASH (Medema et al. 2011; Blin et al. 2017) and
SMURF (Khaldi et al. 2010) are hence set off as the current state-of-the-art tools for
BGC identification. The former is tailored to incorporate both bacterial and fungal
genomes whereas the latter is specific to fungal genomes only. However, SMURF is
only capable of gene prediction and hence must be used with other tools for the
purpose of domain localization. Though these tools have been commonly used for
prediction of BGCs producing various NPs and secondary metabolites in fungi, they
have not been well exploited for siderophores.
As important as the localization of the BGC gene is the prediction of their substrate, particularly to isolate siderophore-producing genes. To facilitate the studies
of “cryptic” NRPS clusters that are not linked to any NP, prediction of substrate
specificity could be conducted (Verne Lee et al. 2015). This could be achieved by
analysis of the substrate-binding adenylation domain ‘A’ in the NRPS module.
Since domain architecture of siderophore-NRPS genes are not unified by a common
pattern, database search approaches have been found useful (Etchegaray et al. 2004;
Komaki et al. 2018; Aleti et al. 2015). Databases are developed based on sequence
10 Bioinformatics Applications in Fungal Siderophores: Omics Implications