Basic information

Indian Medicinal Plants, Phytochemistry And Therapeutics (IMPPAT) database (version 1.1 released on January 25, 2018) captures the following information:

Number of Indian medicinal plants 1742
Number of phytochemicals 9596
Number of associations between plants and phytochemicals 27074
Number of therapeutic uses 1124
Number of associations between plants and therapeutic uses 11514
Number of traditional Indian medicinal formulations 974
Number of associations between plants and medicinal formulations 5069
Number of associations between phytochemicals and human target proteins 48632

Occurrence of phytochemicals across Indian medicinal plants:

In the figure below, we show a histogram of the occurrence of phytochemicals across 1742 Indian medicinal plants in our database. From this figure, it is seen that the majority of phytochemicals are found in less than 5 Indian medicinal plants while only a handful of phytochemicals are found in more than 200 Indian medicinal plants.


Figure 1

Distribution of therapeutic uses across Indian medicinal plants:

In the figure below, we show a histogram of the number of therapeutic uses per Indian medicinal plant in our database. From this figure, it is seen that the majority of Indian medicinal plants have less than 10 documented therapeutic uses while a small fraction of Indian medicinal plants have more than 20 therapeutic uses in our database.


Figure 2

Distribution of physicochemical properties for phytochemicals in our database:

In the figure below, we show the distribution of six physicochemical properties, namely, (A) molecular weight, (B) logP, (C) topological polar surface area, (D) number of hydrogen bond donors. (E) number of hydrogen bond acceptors and (F) number of rotatable bonds, across the 9596 phytochemicals in our database.


Figure 3

Druggability analysis of phytochemicals in our database:

We have also employed cheminformatics approaches to evaluate the drug-likeliness of the phytochemicals in our database using multiple scoring schemes such as RO5, Traffic Lights, GSK’s 4/400, Pfizer’s 3/75, Veber rule and Egan rule. We found a subset of 960 phytochemicals in our database that are potentially druggable in our chemical library based on multiple scoring schemes.

We used the following open source cheminformatic tools were used to compute physicochemical properties and drug-likeliness of phytochemicals:
FAF-Drugs4
RDKit
Open Babel

In the figure below, the horizontal bar plot gives the number of phytochemicals in our database that satisfy different druggability scores. From this figure, it is seen that the majority of our phytochemicals satisfy Veber or Egan rules in comparison to Pfizer’s 3/75 rule or net Traffic Lights value of zero. In the figure below, the vertical bar plot shows the overlap between sets of phytochemicals that satisfy different druggability scores. We found that 960 out of 9596 phytochemicals in our database satisfy all evaluated druggability scores.


Figure 4

In the figure below, we show the classification of the 960 druggable phytochemicals in our database into chemical super-classes obtained from ClassyFire.

Figure 4


Comparison of IMPPAT with earlier databases on phytochemical composition of Indian medicinal plants:

Database IMPPAT Phytochemica Polur et al
Basic statistics
Number of Indian medicinal plants 1742 5 295
Number of phytochemicals 9596 963 1829
Type of associations
Plant-phytochemical associations Yes Yes Yes
Plant-therapeutic use associations Yes No Yes
Plant-medicinal formulation associations Yes No No
Phytochemical-human target protein associations Yes No Yes
Plant part-phytochemical associations No Yes No
Additional Features
Web interface Yes Yes No
Availability of 2D structure of phytochemicals Yes No No
Availability of 3D structure of phytochemicals Yes Yes No
Downloadable structure file formats MOL, MOL2, SDF, PDB & PDBQT MOL2 No
Chemical classification Yes Yes No
Physicochemical properties Yes Yes No
ADMET properties Yes Yes No
Druggability properties Yes No No
Cytoscape network visualization of associations Yes No No
Filter phytochemicals based on physicochemical properties Yes Yes No
Filter phytochemicals based on druggability properties Yes No No
Chemical similarity search within database Yes No No

Comparison of IMPPAT with phytochemical space of Chinese medicinal plants

We have compared the set of 9596 IMPPAT phytochemicals produced by Indian medicinal plants with the set of 10140 TCM-MESH phytochemicals produced by Chinese medicinal plants. By comparing the 9596 IMPPAT phytochemicals with 10140 TCM-Mesh phytochemicals, we find that less than 25%, specifically 2305 phytochemicals, are common between the two databases. Among the 9596 IMPPAT phytochemicals, a subset of 960 phytochemicals were found to be druggable based on multiple druggability scores, namely, RO5, Traffic Lights, GSK’s 4/400, Pfizer’s 3/75, Veber rule and Egan rule. Among the 10140 TCM-Mesh phytochemicals, we found a subset of 972 phytochemicals to be druggable based on multiple druggability scores listed above. We show below the distribution of QEDw scores for the 972 druggable TCM-Mesh phytochemicals which is similar to that for 960 druggable IMPPAT phytochemicals. We also find only a small overlap of 242 phytochemicals between the set of 960 druggable IMPPAT phytochemicals and 972 druggable TCM-Mesh phytochemicals, and thus, phytochemicals from both Indian herbs and Chinese herbs offer extensive opportunity for novel drug discovery.


Figure 6

Eclipta prostataCatharanthus roseusOcimum tenuiflorumCentella asiatica