Anandakumar et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
There is a growing concenter on the importance of medicinal plants in modern health care. The medicinal plant proteins have therapeutic, cosmetic, and other beneficial properties for mankind to treat various diseases. In this regard, 3D models of medicinal plants protein will enhance our knowledge about the identification of new drugs for various diseases and disorders. So, in this work we modeled about 181 various plant proteins of 18 different medicinal plants using molecular modeling techniques (Indian medicinal plants) and made into a dataset IMPPDS. The models are constructed using MODELLER 9v2, with careful manual pair wise alignment. The model templates are obtained based on the sequence similarity (> 40% identity) of known protein structure using BLASTP (Basic Local Alignment Tool for Protein). One or more potential structural templates were selected from PDB (Protein Databank). We used iterative conjugate gradient method for optimizing the model and PROCHECK for validation.
Medicinal PlantMolecular ModelingPairwise alignmentDatasetoptimizingcitationS Anandakumar, V Saravanan, P Shanmughavel (2008)IMPPDS - Indian Medicinal Plants Protein DatasetIntroduction
Many of the medicinal plant proteins have therapeutic, cosmetic, or other beneficial properties (Balunas and Kinghorn, 2005). Three-dimensional structures for most of the Indian medicinal plants protein are limited http://modbase.compbio.ucsf.edu/modbase-cgi/index.cgi, http://www.rcsb.org/pdb. Detailed study on such proteins structure will provide room for innovative applications and help the researchers to have a better insight on the functions and molecular properties of the proteins and to come out with new therapeutic values from corresponding protein http://www.sristi.org/cms/. Also, there exists the obstacle of designing the accurate protein models based on automated computational modeling, chiefly in pair wise alignment (Kolinski and Gront, 2007). So, we performed the modeling procedure in a semi-automated manner with necessary manual pair wise alignment and error interpretation. In this article we described about the dataset of protein model containing 181 different proteins from 18 different medicinal plants.
MethodologyTarget Sequence Retrieval
Names of Medicinal plants protein which have medicinal properties were obtained from the SRISTI http://www.sristi.org/cms/ and sequences were taken from dataset of Swissprot/TrEMBL http://www.expasy.ch/sprot/. All the taken protein sequences were ascertained that the three-dimensional structure of the corresponding proteins were not available in Protein Data Bank and Modbase http://modbase.compbio.ucsf.edu/modbase-cgi/index.cgi, http://www.rcsb.org/pdb. Totally 181 proteins sequence from 18 different medicinal plants were taken.
Template Searching
One or more suitable template proteins for each of the annotated medicinal plants protein were identified based on the sequence similarity (= 40% identity) using BLASTP (Basic Local Alignment Tool for Protein). For each candidate, templates with higher similarity were obtained from Protein Databank http://www.rcsb.org/pdb.
Indian Medicinal Plants Protein Dataset, the dataset snapshot and visual models are shown
Sequence Alignment
Sequence alignment of target and template proteins were derived using the program align2d (alignment package in Modeler). Further, each alignment is manually curated for reduced gaps, insertions and deletions.
Rough Model
A rough 3-D model was constructed for each proteins from the sequence alignment between annotated protein and its corresponding template proteins using MODELLER 9v2 (Eswar et al, 2003) with default parameters of energy minimization value.
Model Refinement
Using the steepest descent and conjugate gradient technique (Sali and Blundell, 1993), the rough constructed models were solvated and subjected to constraint energy minimization. To eliminate bad contacts between protein atoms and structural water molecules the harmonic constraint was set to100 kJ/mol/Å2, applied for all protein atoms.
Evaluation of Refined Model
The refined structure of the models were subjected to a series of tests (Sali et al, 1995), which includes backbone conformation, evaluated by the inspection of the Psi/Phi Ramachandran plot obtained from PROCHECK http://www.biochem.ucl.ac.uk/~roman/procheck/procheck.html analysis, packing quality of the refined structure, investigated by the calculation of PROCHECK Quality Control value and the models can be viewed using PyMol http://www.delsci.com/rel/099/ and RasMol http://www.bernstein-plus-sons.com/software/RasMol_2.7.2.1.1/ README.html.
Usefullness And Biological Community
This dataset contains 181 protein models [figure 1] from eighteen different various Indian Medicinal plants such as Bauhinia variegata, Camellia sinensis, Allium cepa, Allium ascalonicum, Hibiscus rosasinensis, Punica granatum, Brassica juncea, Mangifera indica, Euphorbia pulcherrima, Aloe vera, Crotalaria juncea, Coriandrum sativum, Cuscuta reflexa, CyNodon dactylon, Citrus sinensis, Citrus reticulata, Citrus clementina and Citrus limon. Each protein models were curated and verified by manually. Detailed study on such modelled Indian medicinal plants protein structure will provide room for innovative applications and help the researchers to have a better insight on the functions and molecular properties of the proteins and to come out with new therapeutic values.
ReferencesBalunasMJKinghornAD2005Drug discovery from medicinal plantsLife Sci8431441EswarN2003Tools for comparative protein structure modeling and analysisNucleic Acids Res3133753380KolinskiAGrontD2007Comparative modeling without implicit sequence alignments BioinformaticsBioinformatics2325222527SaliABlundellTL1993Comparative protein modelling by satisfaction of spatial restraintsJ Mol Biol234779815SaliA1995Evaluation of comparative protein modeling by MODELLERProteins23318326