A knowledge-driven protocol for prediction of proteins of interest with an emphasis on biosynthetic pathways

Abstract

This protocol describes a stepwise process to identify proteins of interest from a query proteome derived from NGS data. We implemented this protocol on Moringa oleifera transcriptome to identify proteins involved in secondary metabolite and vitamin biosynthesis and ion transport. This knowledge-driven protocol identifies proteins using an integrated approach involving sensitive sequence search and evolutionary relationships. We make use of functionally important residues (FIR) specific for the query protein family identified through its homologous sequences and literature. We screen protein hits based on the clustering with true homologues through phylogenetic tree reconstruction complemented with the FIR mapping. The protocol was validated for the protein hits through qRT-PCR and transcriptome quantification. Our protocol demonstrated a higher specificity as compared to other methods, particularly in distinguishing cross-family hits. This protocol was effective in transcriptome data analysis of M. oleifera as described in Pasha et al. Knowledge-driven protocol to identify secondary metabolite synthesizing protein in a highly specific manner. Use of functionally important residues for screening of true hits. Beneficial for metabolite pathway reconstruction in any (species, metagenomics) NGS data.

Genome sequencing of turmeric provides evolutionary insights into its medicinal properties

“…The ten enzymes involved in curcuminoid biosynthesis pathway were analyzed to elucidate their origin by remote homology finding and to examine their phylogenetic position with respect to the genes from other plant species (see "Methods"). These enzymes were identified in the coding gene set of C. longa using CAPS_protocol 81 . Among the ten enzymes, HCT had all the phylogenetically closer orthologs (transferases) from angiosperm plant division.…”

“…Further, the ten enzymes involved in curcuminoid biosynthesis pathway (Fig. 3a) were searched, and identified in the proteome sequences of C. longa using CAPS_protocol 81 . EC numbers or NCBI accession numbers of these ten enzymes (Supplementary Table 22) were used for homolog identification for each enzyme in UniProt database 134 , and the top hits that were found in UniProt database were retained.…”

“…In order to identify the true homologs, the functionally important residues (FIR)-binding site and active site amino acid residues for each enzyme (Supplementary Table 23) were detected from UniProt 134 database, and sequences that did not contain those residues were removed from the alignments. These filtered alignments were queried against the proteome sequences of C. longa using PSI-BLAST 141 with e-value of 10 −5 , inclusion threshold of 10 −5 , query coverage ≥70%, sequence identity ≥40%, and 2 iterations, as used in CAPS_protocol 81 . The PSI-BLAST hits were again searched for the presence of FIRs, and the best identical hits were retained.…”

Exploring the medicinally important secondary metabolites landscape through the lens of transcriptome data in fenugreek (Trigonella foenum graecum L.)

“…We considered trigonelline and glycine-betaine from the alkaloid class found in fenugreek. We pursued identification of enzymes involved in their biosynthesis by adopting the strategy described in Joshi et al 38 .…”

“…The genes involved in synthesis of secondary metabolites belonging to alkaloids, saponins, volatiles and flavonoids, present in T. foenum-graecum were identified using strategy adapted from Joshi et al 38 . The pathway information and related functional annotation was extracted from multiple sources such as PlantCyc MetaCyc, KEGG and literature 11 , 30 , 39 – 42 .…”