Skip to main content
NCBI home page
3 Biotech logoLink to 3 Biotech
. 2021 Oct 18;11(11):463. doi: 10.1007/s13205-021-03001-5

Quantification of adulteration in traded ayurvedic raw drugs employing machine learning approaches with DNA barcode database

Suma Arun Dev 1,✉,#, Remya Unnikrishnan 1,5,#, R Jayaraj 3, P Sujanapal 2, V Anitha 4
PMCID: PMC8523608  PMID: 34745814

Abstract

Adulteration of expensive raw drugs with inferior taxa has become a routine practice, conceding the quality and safety of derived herbal products. In this regard, the study addresses the development of an integrated approach encompassing DNA barcode and HPTLC fingerprinting to authenticate chiefly traded ayurvedic raw drugs in south India [viz. Saraca asoca (Roxb.) Willd., Terminalia arjuna (Roxb. ex DC.) Wight and Arn., Sida alnifolia L. and Desmodium gangeticum (L.) DC.] from its adulterants. Consortium of Barcode of Life (CBOL) recommended DNA barcode gene regions viz. nuclear ribosomal—Internal Transcribed Spacer (nrDNA-ITS), maturase K (matK), ribulose-1,5-bisphosphate carboxylase/oxygenase large subunit (rbcL) and psbA-trnH spacer regions along with HPTLC profiling were experimented and a reference database was created. Further, an integrated analytical approach employing genetic distance-based Maximum Likelihood phylogenetic tree and Artificial Intelligence (AI)based Machine Learning Algorithms (MLA)-Waikato Environment for Knowledge Analysis (WEKA) and Barcoding with Logic (BLOG) were employed to prove efficacy of DNA barcode tool. Even though, among the four barcodes, psbA-trnH (S. alnifolia and its adulterants, T. arjuna and its adulterants) or ITS region (S. asoca and its adulterants, D. gangeticum and its adulterants) showed highest inter specific divergences in the selected Biological Reference Materials (BRMs), rbcL or matK barcode regions alone were successful for authentication of traded samples. The automated species identification techniques, WEKA and BLOG, experimented for the first time in India for raw drug validation, could achieve rapid and precise identification. A national certification agency for raw drug authentication employing an integrated approach involving a DNA barcoding tool along with standard organoleptic and analytical methods can strengthen and ensure safety and quality of herbal medicines in India.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13205-021-03001-5.

Keywords: Machine Learning Algorithm, Artificial Intelligence, DNA barcoding, Ayurvedic raw drug

Introduction

India is renowned for its traditional medicine systems (Ayurveda, Siddha, Unani, Homeo, etc.) and is the largest producer of medicinal herbs in the world (Ganesan et al. 2016). As per the recent report of Food and Agriculture Organization (FAO) and United Nations Environmental Programme (UNEP), more than 28,000 plant species are being utilized worldwide for various medicinal purposes (FAO and UNEP 2020). In India, 8000 plant species have been utilized in various systems of medicine with approximately 25,000 effective herbal formulations and 1200 actively traded species (Goraya and Ved 2017; Gautam et al. 2020). This huge demand for herbal drugs invariably leads to the upsurge of herbal industries in India. Commercial demand of herbal drugs has been recently valued at INR 300 billion in 2018, which is more than double as compared to preceding years (CISION PR News Wire 2020). Although the use of herbal medicines has become commercialised, majority of plants are still harvested from the wild. The growing demand and limited supply exert a heavy pressure on the available natural resources (Goraya and Ved 2017; Joshi et al. 2017). Consequently, overexploitation, unscientific extraction and limited supply have instigated adulteration and substitution of potential raw drugs (Newmaster et al. 2013; Srirama et al. 2017; Simmler et al. 2018). Recently, raw drug adulteration has become a pressing concern in herbal industries (Walker and AppleQuist 2012; Ouarghidi et al. 2013; Techen et al. 2014; Mishra et al. 2016). The consumer’s faith on herbal medicines is in the phase of decline due to extremities in adulteration/substitution and ensuing consequences. The percentage of adulterated products varies among countries and the highest percentage of herbal product adulteration was reported from Brazil, followed by India (Ichim 2019).

To ensure the quality of herbal medicines, WHO pharmacopeia and Ayurvedic Pharmacopoeia of India (API) have implemented certain criteria for proper identification of plant species and assessment using potent chemical markers to guarantee the quality of herbal medicines (WHO 2011; Joshi et al. 2017). API is a legally binding document describing the quality, purity, and strength of selected drugs that are manufactured, distributed and sold by the licensed manufacturers in India (Joshi et al. 2017). The quality assurance is mandatory for the extensive usage and acceptability of herbal formulations in complementary and alternative medicine practicing countries (WHO 2011). Although, there are several recommended methods such as morphological, chemical/biochemical and organoleptic, it is extremely difficult to confirm authenticity of traded dried form of raw drugs (Coghlan et al. 2012). The lack of taxonomic expertise and phenotypic plasticity also complicate the identification of raw drug species from the wild. In many instances, the quality and quantity estimation of potent herbal medicines are performed using chemical finger printing methods and Thin Layer Chromatography (TLC) is the basic identification test in most of the Pharmacopeia. Recently, TLC is replaced by High Performance Thin Layer Chromatography (HPTLC) owing to its accuracy and reproducibility (Upton 2010). HPTLC finger printing has been successfully employed for the rapid identification of Apiaceae species, adulteration in Aristolochia herbal materials, profiling of Tinospora species, chemical standardisation of Sida species, authentication of Taraxacum officinale and authentication of Echniacea herbal products, among others (Khatoon et al. 2005; Cortes et al. 2014; Dechbumroong et al. 2018; Raclariu et al. 2018; Shawky and Abou 2018; Parveen et al. 2020). However, various external environmental factors such as age of plant, storage conditions and also type of the plant parts used can significantly influence the chemical fingerprints (Liu et al. 2011; Kaur et al. 2016). Recently, DNA barcode based plant identification has gained profound consideration as a powerful method in herbal pharmacovigilance research, mostly for quality assurance in conjunction with analytical/hyphenated methods (WHO 2011; De Boer et al. 2015; Palhares et al. 2015; Kumar et al. 2016a, 2016b; Seethapathy et al. 2018). DNA barcode analysis has revealed a wide range of discrepancies in the claimed composition of various herbal products and authenticity of raw market samples viz. saw palmetto dietary supplements, ginkgo products, black cohosh, herbal teas, ginseng and raw drugs like Aswagandha, Cinnamomum verum, Cassia, Myristica fragrans, Phyllanthus, Sida and Santalum album (Stoeckle et al. 2011; Baker et al. 2012; Wallace et al. 2012; Dev et al. 2013; Little and Jeanson 2013; Newmaster et al. 2013). Computer aided Artificial Intelligence (AI) based Machine Learning Algorithm (MLA) has been reported as an accurate means for species authentication (More et al. 2016; He et al. 2018). Recently, supervised Machine Learning Algorithm, WEKA and BLOG were utilised for the authentication of recently diverged, morphologically similar timber species (Dalbergia and Pterocarpus) (Hartvig et al. 2015; He et al. 2018) and raw drug authentication of traded medicinal plants in herbal industries (Hartvig et al. 2015; He et al. 2018, 2019; Unnikrishnan et al. 2021). Ability to analyse huge data is one of the major advantages of Artificial Intelligence, which offers quite a large platform for rapid authentication of raw drug samples at the industrial scale, which subsequently reduces the human labour (Hartvig et al. 2015).

In this backdrop, the present study addresses the development and validation of an integrated approach involving DNA barcoding as well as High Performance Thin Layer Chromatography (HPTLC) fingerprinting to authenticate majorly traded selected ayurvedic raw drugs [Saraca asoca (Roxb.) Willd., Terminalia arjuna (Roxb. ex DC.) Wight & Arn., Sida alnifolia L. and Desmodium gangeticum (L.) DC.], identified through market and industrial surveys in south India as part of earlier research projects. The study also compares the efficacy of Machine Learning Algorithm (WEKA and BLOG) over the genetic distance based approaches to analyse the barcode sequence data and hierarchical clustering for chemical finger prints to ascertain the extent of adulteration.

Materials and methods

Collection of authenticated biological reference material (BRM)

The highly traded raw drugs and its market adulterants were identified through a preliminary survey. Mature stem, leaf and flower samples of Saraca asoca and its adulterants [Polyalthia longifolia (Sonn.) Thwaites, Polyalthia coffeoides (Thwaites) Hook.f. and Thomson, Shorea roxburghii G.Don], Terminalia arjuna and its adulterants [Lagerstroemia speciosa (L.) Pers., Lagestroemia microcarpa Hance], Sida alnifolia and its adulterants (Sida acuta Burm.f., Sida cordifolia L., Sida rhombifolia L., Urena lobata L.), Desmodium gangeticum and its adulterants [Desmodium triangulare (Retz.) Merr., Desmodium triquetrum (L.) DC., Desmodium pulchellum (L.) Benth., Desmodium velutinum (Willd.) DC.] (n = 68, 17 species × 4 accessions) were collected from different geographic locations of its distribution zones in south India for development of the DNA barcode database (Fig. 1). Medicinally important parts (bark of S. asoca and its adulterants as well as T. arjuna and its adulterants, whole plants of S. alnifolia and its adulterants as well as D. gangeticum and its adulterants) were collected with two individuals each (n = 34) for HPTLC analysis. GPS coordinates of the locations and details of the collected plant samples are provided in Supplementary table S1. Multiple accessions were collected and specimens were stored in silica gel for further use. The voucher specimens were deposited in the KFRI herbarium.

Fig. 1.

Fig. 1

Open in a new tab

Four selected taxa (Desmodium gangeticum, Saraca asoca, Terminalia arjuna, Sida alnifolia) of ayurvedic raw drugs

Collection of traded samples

Traded samples such as dried plant parts such as leaf, stem, root, and bark were collected from the selected authorised dealers of raw drugs and major herbal industries in south India (n = 60) (Supplementary table S2). About 100 g of each raw drug (available in extremely dried and shredded form) was purchased from authorized shops to check adulteration in the traded samples. Each collected raw drug sample was given Herbal Authentication Service Code (HAS) with details of location. To avoid the chances of mixing up, care was taken from collection to final data analysis. Most of the procured raw drugs had not retained any morphological features of the original plant species. Raw drugs of all the selected species were able to purchase from herbal markets except T. arjuna.

DNA extraction

Genomic DNA extraction was performed using modified Cetyl Trimethyl Ammonium Bromide (CTAB) method (Doebley et al. 1990) as well as DNeasy Plant Mini Kit for difficult samples according to manufacturer’s protocol (Qiagen, USA). Total genomic DNA was also extracted from the collected raw drug samples. DNA samples were stored in the deep freezer at − 20 °C until further use. The samples were separated on 1.5 per cent agarose gel and stained in ethidium bromide and visualised under UV transilluminator to check the quality. It was quantified using a nanospectrophotometer (Nanodrop Fisher Thermo., USA).

PCR amplification

The taxonomically authenticated Biological Reference Materials (BRM) were used to develop species specific barcodes for four standard barcode gene regions (rbcL, matK, ITS and psbA-trnH). The barcode regions, primer sequences and PCR conditions used are provided in Supplementary table S3.

Working concentration of genomic DNA was prepared by diluting the stock solution at a concentration 25 ng/μL. 25 μL of PCR reaction mixture comprised of 2.5 μL PCR buffer at 1× (supplied with 10× concentration), 1 μL each of forward and reverse primers (5 pmol), 2.5 μL of dNTPs from 10 mM stock, 2 U/25 μL of Taq-polymerase, 1 μL template DNA with the concentration of 25 ng/μL and the final volume of the PCR reaction mixture was made up to 25 μL with sterile distilled water. PCR reaction was performed with the following conditions, initial denaturation of 5 min at 94 °C, cycle denaturation of 1 min at 94 °C, cycle annealing of 1 min at 60 °C (rbcL and ITS), 58 °C (psbA-trnH) and 50 °C (matK) and cycle extension of 1 min at 72 °C for 35 cycles and a final extension at 72 °C for 10 min. PCR products were resolved by 2 per cent agarose. Electrophoresis was performed on agarose gel by applying constant voltage to resolve the products and documented with Alpha Imager (Alpha Innotech, USA).

Sequencing of PCR products

PCR reaction was scaled up to 50 μL for the elution of PCR products by Nucleospin gel and PCR clean up kit as per the manufacturer’s protocol (Machery-Nagel, U.S.A.). DNA sequencing was performed for the eluted PCR products in both forward and reverse directions employing Sanger’s dideoxy chemistry.

HPTLC analysis

CAMAG Linomat 5 with twin plate chamber and CAMAG TLC scanner instrument programmed through Win CATS software were used for HPTLC finger printing. Medicinally important parts (bark of S. asoca and its adulterants as well as T. arjuna and its adulterants, whole plants of S. alnifolia and its adulterants as well as D. gangeticum and its adulterants) with two individuals each (n = 34) were finely powdered. Ten gram of powder accurately weighed from each sample, was extracted using solvents. Methanol extraction was carried out for S. alnifolia, S. asoca, D. gangeticum and its adulterants whereas chloroform extraction was performed for T. arjuna. Extracts were filtered and concentrated under reduced pressure and made upto 10 mL. Stationary phase of aluminium TLC plates was pre-coated with silica gel 60 F254 of 0.2 mm thickness and mobile phase for each set was standardised (Supplementary table 4). Samples were visualised at 366 nm. The plates were derivatized with anisaldehyde sulphuric acid reagent for band visualization of phenolic components at 366 nm.

HPTLC banding profile of BRM was documented at 366 nm under UV light. Chemical profiles of samples were analysed according to their RF values (Retention factor). Dendrogram was constructed using SPSS v.16.0 (SPSS Inc 2007) with nearest neighbour, adopting euclidean distances, which revealed the relationships between each species as per their phytochemical constituents. Owing to the variations in HPTLC fingerprints among geographical accessions of species, the technique could not be further considered for market sample validation.

Sequence data analysis

Raw chromatograms were edited and trimmed using BioEdit software (Hall 1999). The edited sequences were aligned using Clustal W (Thompson et al. 1997)  and submitted to GenBank (http://www.ncbi.nlm.nih.gov/genbank/) (Supplementary table S5). Homology searches were performed using BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) to confirm the identity of the sequences.

For pair-wise genetic distance (PWG) method, genetic pair-wise distances (interspecific as well as intraspecific distances) were determined by MEGA v.6.0 using Kimura two-parameter distance model (K2P) adopting complete deletion option (Tamura et al. 2013). The interspecific divergence between species was calculated using three parameters; (i) average interspecific distance, (ii) average theta prime (θ’) and (iii) minimum interspecific distances. Intraspecific parameters; (iv) average intraspecific distance, (v) theta (θ) and (vi) coalescent depth were also calculated to characterize intraspecific divergences (Meyer and Paulay 2005). DNA barcoding gaps were calculated by comparing intra and interspecific genetic distances (Meyer and Paulay 2005; Meier et al. 2006). Significance of barcoding gap was assessed using Wilcoxon matched pairs signed rank test in SPSS v.16.0 (SPSS Inc 2007). Dendrogram was constructed using developed barcode sequences for authentication of traded samples based on BRM, with 1000 bootstrap using MEGAv.0.7 adopting Kimura 2 model (Kumar et al. 2016a, 2016b).

Machine learning algorithm (MLA)-based analysis (WEKA and BLOG)

MLA’s, WEKA is a GUI tool, which allows loading a data set, running an algorithm and designing an experiment. In WEKA, DNA barcoding analysis was performed with a reference data set composed of DNA sequences of known species (BRM) and query data set comprising sequences of unknown species (market samples), the function-based method Support Vector Machines (SMO) (Suykens and Vandewalle 1999), the rule-based RIPPER (Jrip) (Shahzad et al. 2013), the decision tree C4.5 (J48) (Quinlan 1996) and the Bayesian-based method Naive Bayes (Lewis 1998) were tested on DNA barcodes with ten-fold cross validation. The “.fasta” files of barcode sequences were converted to “.arff ” format using “Fasta2Weka” programme for analysis in WEKA (Weitschek et al. 2014). All four classification methods in WEKA were run with four barcode primer sequences of BRM. Best classifier was selected according to their efficiency in species discrimination. Using the best classifier, sequences from traded market samples were further analysed along with BRM sequence database.

BLOG 2.0 (Barcoding with logic) is specifically developed to handle the DNA barcode data, which provides a character-based method, to identify the species using key diagnostic nucleotides. Input parameters for feature selection were given as, a maximum number of 35 features chosen (‘BETA = 35’), a maximum of 200 iterations (‘GRASPITER = 200’), and a maximum time of 500 min for analysis (‘GRASPSECS = 30,000’). BLOG uses reference data set as training data to compute the classification rules, which was applied to test data of unknown market samples. BLOG produces a set of logic rules for the data set (BRM sequence) which can be utilized to identify the test species (market samples).

Results

DNA barcode analysis

All the analysed barcode regions (ITS, psbA-trnH, matK and rbcL) showed 100 per cent PCR amplification in all the Biological Reference Materials (BRM) selected (Supplementary Fig. 1). Intra- and inter-specific genetic divergences of the selected BRM sequences analysed using the four barcode regions showed interspecific as well as intra-specific divergences (Table 1). Among these four barcode regions, either psbA-trnH (S. alnifolia and its adulterants, T. arjuna and its adulterants) or ITS region (S. asoca and its adulterants, D. gangeticum and its adulterants) showed highest inter and intra-specific divergences in the selected BRMs. Wilcoxon’s signed rank test performed to test the significance of interspecific divergences in the barcode regions showed significant values (p value ≤ 0.05).

Table 1.

Evaluation of the four DNA barcode regions used in this study

Species Average intraspecific divergence Average interspecific divergence
rbcL matK psbA-trnH ITS rbcL matK psbA-trnH ITS
Desmodium gangeticum + adulterants 0 0 0.0013 ± 0.0008 0 0.00463 ± 0.001 0.0151 ± 0.0028 0.0278 ± 0.0049 0.1722 ± 0.1061
Saraca asoca + adulterants 0 0 0 0 0.2064 ± 0.0761 0.0541 ± 0.104 0.1944 ± 0.208 0.321 ± 0.1297
Sida alnifolia + adulterants 0 0 0.0006 ± .0003 0.0008 ± 0.002 0.00083 ± 0.00054 0.0033 ± 0.00096 0.0812 ± 0.0196 0.0207 ± 0.0028
Terminalia arjuna + adulterants 0 0 0 0 0.066 ± 0.0024 0.040 ± 0.00487 2.65 ± 1.99 0.068 ± 0.0109
Open in a new tab

DNA barcode analysis of the extremely shredded traded specimens showed successful PCR amplification for rbcL barcode region except for D. gangeticum. Thus, rbcL barcode region alone was adopted for authentication of traded samples of Saraca asoca and Sida alnifolia and matK barcode region was opted for D. gangeticum. The barcode sequences developed from the traded samples clustered along with the respective sequences of BRM samples in the dendrogram. The phylogenetic tree generated based on these sequences showed a clear clustering of traded samples with those of BRM [D. gangeticum (Fig. 2), S. asoca (Fig. 3), S. alnifolia (Fig. 4)], thereby, presence of adulterants could be clearly identified. Twenty traded samples of S. asoca were clustered along with the BRM of P. longifolia. Out of 20 traded samples of S. alnifolia, 15 samples clustered with S. cordifolia and 5 with S. acuta. Traded samples of D. gangeticum showed least similarity with any of the BRM and formed a clade sister to D. gangeticum in the ML tree.

Fig. 2.

Fig. 2

Open in a new tab

Maximum Likelihood tree (ML) of market samples along with BRM (D. gangeticum)

Fig. 3.

Fig. 3

Open in a new tab

Maximum Likelihood tree (ML) of market samples along with BRM (S. asoca)

Fig. 4.

Fig. 4

Open in a new tab

Maximum Likelihood tree (ML) of market samples along with BRM (S. alnifolia)

Machine learning algorithm (WEKA and BLOG)

Results similar to the distance base approach were generated in the MLA analysis. The identification rates of four machine learning algorithms in the selected raw drugs are provided in Table 2. Performance success of different machine learning algorithms also varied for BRM and market samples (Table 3). Best performance was shown by SMO and J48 with 100 per cent accuracy in authenticating the test data of all traded samples, though Jrip and Naive bayes were also successful in combinations with SMO and J48 (Fig. 5). MLA results substantiated the phylogenetic tree-based analysis.

Table 2.

Species identification rates as percentage (correctly identified/misidentified/not identified) for the BRM using Machine Learning Algorithm (WEKA and BLOG)

Barcoding primers WEKA BLOG
Naive Bayes SMO J48 Jrip
Desmodium gangeticum + adulterants
 rbcL 30/70/0 100/0/0 70/30/0 65/35/0 100/0/0
 matK 100/0/0 100/0/0 100/0/0 95/5/0 100/0/0
 psbA-trnH 100/0/0 100/0/0 100/0/0 100/0/0 100/0/0
 ITS 95/5/0 100/0/0 95/5/0 90/10/0 55/45/0
Saraca asoca + adulterants
 rbcL 100/0/0 100/0/0 100/0/0 100/0/0 100/0/0
 matK 60/40/0 100/0/0 100/0/0 63/37/0 100/0/0
 psbA-trnH 100/0/0 100/0/0 100/0/0 100/0/0 100/0/0
 ITS 100/0/0 100/0/0 100/0/0 100/0/0 100/0/0
Sida alnifolia+adulterants
 rbcL 40/60/0 100/0/0 75/25/0 100/0/0 100/0/0
 matK 0/100/0 100/0/0 75/25/0 100/0/0 100/0/0
 psbA-trnH 20/0/0 20/0/0 20/0/0 100/0/0 100/0/0
 ITS 90/10/0 100/0/0 95/5/0 95/5/0 100/0/0
Open in a new tab

Table 3.

Species identification rate of market samples along with BRM

BRM + Traded samples Barcode region WEKA BLOG
Naive Bayes SMO J48 Jrip
Desmodium gangeticum+traded sample+adulterants matK 34/66/0 100/0/0 95/5/0 – 100/0/0
Saraca asoca+adulterant+traded samples rbcL 100/0/0 100/0/0 100/0/0 100/0/0 100/0/0
Sida alnifolia+adulterants+traded samples rbcL – 100/0/0 100/0/0 – 100/0/0
Open in a new tab

Fig. 5.

Fig. 5

Open in a new tab

The confusion matrix (output of WEKA) showing identification rate of market samples along with BRM (a. Desmodium gangeticum, b. Saraca asoca, c. Sida alnifolia)

BLOG successfully identified the test data and generated logic formula for the identification of unknown market samples (Table 2). The identification rate of train data (BRM) and test data (market samples) are provided in the table (Table 3). BLOG showed higher identification rate in rbcL and matK barcode regions to authenticate the market samples or unknown samples. Logic rules derived from the train data (BRM samples) were utilized for the identification of test data (Market samples). BLOG generated logic formula was used to identify the presence of adulteration in market samples of selected raw drugs.

HPTLC analysis

Dendrogram generated using RF values was used to analyse the banding pattern of BRM samples (Supplementary Fig. 2). Accessions of a species belonging to different geographical locations showed intraspecific variations in their banding patterns. For example, chemical fingerprints of D. velutinum collected from two different geographical locations clustered separately. Interspecific similarities in HPLTC profiles were also evident as in the case of D. velutinum which was collected from Aralam, showed similarity to that of D. gangeticum and D. velutinum samples from Athirapilly showed similarity to D. triquetrum. Similarly, P. longifolia collected from KFRI showed similarity to P. coffeoides and P. longifolia from Palakkad showed similarity to S. roxburghii. Further, S. alnifolia showed similarity to S. rhombifolia, while S. cordifolia showed similarity to U. lobata (Figs. 6, 7). Therefore, HPTLC fingerprinting could not be further used as a species authentication tool because of the inconsistencies in the fingerprints generated.

Fig. 6.

Fig. 6

Open in a new tab

Dendrogram constructed using HPTLC banding pattern (RF value) (a. Saraca asoca and its adulterants, b. Desmodium gangeticum and its adulterants)

Fig. 7.

Fig. 7

Open in a new tab

Dendrogram constructed using HPTLC banding pattern (RF value) (c. Sida alnifolia and its adulterants, d. Terminalia arjuna and its adulterants)

Discussion

Authentic herbal drugs play a crucial role in determining the quality, safety and efficacy of herbal formulations. Therefore, to guarantee the quality of herbal medicines, WHO pharmacopeia or international pharmacopeia implemented certain criteria for appropriate identification of the plant species and quality valuation of potent phytochemical principles (Palhares et al. 2015). Yet adulteration/substitution is a prevailing issue in herbal industries. Though, herbal medicine has gained much popularity in India and all around the world, proper certification is yet to be in place. Herbal medicine once formulated is easily available to the public through various portals, where no mention of any clinical trials or authenticity is cited. This demands a proper industry standard and organized public BRM library for herbal products. The development of voucher Biological Reference Material (BRM) is a critical part of raw drug authentication process.

Even though vast array of physical, chemical (analytical), biochemical, anatomical and organoleptic methods are available, DNA barcode-based method of plant identification has comparatively been considered as the most powerful and potential standard in herbal pharmacovigilance research (Palhares et al. 2014). In India, 31 per cent of herbal drug adulteration and wide range of incongruences between claimed and identified species composition have been reported through DNA barcode-based authentication studies (Moustafa et al. 2016). Recently, an integrated approach of two or more advanced techniques was employed for species authentication. This multi-tier approach of DNA barcode along with HPTLC, NMR or HPLC was utilized for quality assurance and species authentication of popular species such as Hamamelis virginiana, Matricaria recutita, Maytenus ilicifolia, Mikania glomerata, Panax ginseng, Passiflora incarnata, Peumus boldus and Valeriana officinalis (Moustafa et al. 2016), Garcinia species, Terminalia species and Saraca asoca (Nithaniyal and Parani. 2016; Mishra et al. 2017; Santhosh Kumar et al. 2018; Seethapathy et al. 2018). However, studies have also reported the discrepancy of chemical markers in delineating medicinal plants, owing to variation with age of plant, dioecious nature and environmental heterogeneity (Liu et al. 2011; Cao et al. 2012; Joshi et al. 2016; Kaur et al. 2016; Upton et al. 2019). Analytical methods also failed to differentiate closely related species containing similar chemical constituents in some instances (Bajpai et al. 2017).

An integrated approach involving CBOL recommended barcode regions such as rbcL, matK, psbA-trnH, ITS as well as HPTLC profiling was investigated in the present study to create a reference database for majorly traded raw drugs and its market adulterants in India. However, accessions of species collected from different geographic locations of unknown age, showed variations in the HPTLC fingerprints which restricted their further use in raw drug authentication. Similarly, accessions of species such as Artemisia japonica, Cinnamomum glaucescens and Cymbopogon distans collected from different geographical locations also showed variation in their chemical pattern (Joshi et al. 2017). Moreover, sex specific discrepancies in the phytochemical constituents were also reported in Tinosporia cordifolia and Dioscorea (Chase et al. 2007; Paul et al. 2020).

Even though, DNA barcode regions like ITS showed promising results in the case of D. gangeticum, T. arjuna, S. asoca and psbA-trnH in S. alnifolia, high number of indels along with huge interspecific variation failed to provide consistent bidirectional unambiguous sequencing reads which limited their utility for authentication (CBOL 2009; Hollingsworth et al. 2009; Roy et al. 2010; Kool et al. 2012). Further, degraded DNA obtained from the market samples with impurities like fungal contamination and presence of extraneous matters hindered the primer annealing and subsequent PCR amplification of ITS/psbA-trnH barcode gene regions. Similar hurdles were also reported in the traded samples of medicinal plants in Morocco, India and Brazil (Srirama et al. 2010; Santhosh Kumar et al. 2018; Parveen et al. 2020). Consequently, rbcL and matK barcode sequence database which was discriminant enough to identify adulterants in all the cases were used to validate the traded market raw drugs.

Ability to analyse large data is one of the major advantages of Artificial Intelligence. It also provides quite a large platform for rapid authentication of raw drug samples, which subsequently reduces human labour (He et al. 2018). MLA’s created specifically for DNA barcoding can be effectively used for species authentication in herbal industries, timber industries and for biodiversity conservation (He et al. 2018; Heinrich et al. 2018). Recent studies authenticated the efficiency of Machine Learning Approaches (WEKA and BLOG) for species delineation of even recently diverged species, and morphologically similar timber species such as Dalbergia and Pterocarpus as compared to conventional barcode sequence analysis methods [Maximum Likelihood (ML), BLAST and nearest-neighbour] (He et al. 2018, 2019). The present study demonstrated the efficiency of MLA (WEKA and BLOG) for DNA barcode sequence data analysis to confirm the presence of adulterants in commercially traded samples of S. asoca, S. alnifolia and D. gangeticum. Traded samples of S. asoca showed similarity with Polyalthia longifolia, S. alnifolia with S. cordifolia/S. acuta and D. gangeticum traded samples showed no similarity with the developed reference database. MLA barcode data analysis thus can be recommended as a powerful, accurate and rapid means over genetic distance-based methods to discriminate original raw drugs from its adulterants based on the case studies.

DNA barcode authentication analysis revealed the presence of adulteration in the traded market samples of the studied species. Market samples of S. asoca were adulterated with P. longifolia, S. alnifolia with S. cordifolia/S. acuta and D. gangeticum with other similar species of Desmodium. Similar vernacular names, presence of potent chemicals, morphological similarities as well as overlapping species distributions are considered as primary reasons for adulteration (Srirama et al. 2010). In the studied species, Sida alnifolia, S. cordifolia and S. acuta are locally known as “bala” and contain the potent chemicals like ephedrine, vasicinone and rutin (Dhalwal and Shinde 2010; Chaves et al 2013). Similarly, S. asoca and P. longifolia known as ‘asoka’, contain caffeic acid and ellagic acid in common (Sharma et al. 2018). Similar issues of adulteration were reported wherein Myristica fragrans was adulterated with M. malabarica, Cinnamomum verum with C. cassia/C. malabatrum (Swetha et al. 2014, 2017). Adverse consequences of herbal drug adulteration were reported from the countries like Australia, Japan, China and regions like Taiwan, where chronic use of adulterated Artistolochia fangchi, led to the death of patients owing to renal failure (Michl et al. 2013; Jadot et al. 2017). Similarly, US Food and Drug Administration banned Piper methysticum containing products in Germany, Switzerland, France, Canada and UK due to health issues related to hepatitis, cirrhosis and liver failure (U.S. Food and Drug Administration 2002).

In India, there are no proper guidelines to coordinate and maintain the information related to collection, supply, trade and consumption of botanicals (Goraya and Ved 2017; Gauthum et al. 2020). Herbal sector in India is a large network involving various government agencies, stake holders, herbal healers and traders. Though all of them are linked in the trade network, authenticity of the raw drugs extracted from wild through formal and informal sectors is mostly ambiguous. Recently, to bring in more transparency, National Medicinal Plant Board (NMPB), Govt. of India, has launched an online virtual platform ‘www.e-charak.in’ and a mobile application ‘e-charak’ with a view to provide online market portal and to ensure the transparent trade of medicinal plants among producers and purchasers in the country (https://www.nmpb.nic.in/content/e-charak-online-market-place-medicinal-plants).

Conclusions

Our study could demonstrate the effectiveness of DNA barcode analysis through Artificial Intelligence (AI) platform to authenticate ayurvedic raw drug species in addition to the recommended standard organoleptic and analytical methods. DNA barcode reference library once created and deposited in the public database  can be further accessed for species identification of unknown samples, whenever required for certification purposes. British pharmacopeia is the first agency to globally implement DNA barcoding tool for authentication, given its ability to identify the source of herbal products accurately (Heinrich et al. 2018; Sgamma et al. 2018). Though herbal medicine has gained much popularity in India, proper certification procedures and agencies are yet to be established. It is therefore imperative to bring forth a statutory body to monitor proper collection, processing, certification and sale of raw drugs in the country. Substitution of authentic species with species of similar therapeutic effects based on the ancient ayurvedic scripts can alleviate the extraction pressure and conservation measures can augment the depletion of wild resources.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOC 147 KB) (147KB, doc)
13205_2021_3001_MOESM2_ESM.jpg (87.8KB, jpg)

Supplementary file2 (JPG 88 KB) Agarose gel electrophoresis (a) (M- 100 bp ladder, matK, rbcL, psbA-trnH) (b) (M-100 bp ladder, ITS)

13205_2021_3001_MOESM3_ESM.jpg (186.1KB, jpg)

Supplementary file3 (JPG 186 KB) HPTLC banding pattern (a. S. alnifolia and its adulterants, b. D. gangeticum and its adulterants, c.T. arjuna and its adulterants, d. S. asoca and its adulterants)

Acknowledgements

We are thankful to the State Forest Department of Kerala, for their permission to collect wild samples. The financial support received from National Medicinal Plant Board (NMPB), Ministry of AYUSH, Government of India (No. F.No. Z.18017/187/CSS/R&D/KE-01/2016-17-NMPB-IV A) is gratefully acknowledged.

Author contributions

SAD, JR: Fund acquisition, conceptualization, methodology and designing the experiments, writing, editing and finalization of manuscript. RU: Conducted wet lab experiments as part of doctoral research, data analysis and manuscript writing. SP, AV: Sample collection and manuscript writing.

Declarations

Conflict of interests

The authors declare no conflict of interests.

Footnotes

Suma Arun Dev and Remya Unnikrishnan have contributed equally.

References

  1. Bajpai V, Kumar S, Singh A, Singh J, Negi MPS, Bag SK, Kumar B. Chemometric based identification and validation of specific chemical markers for geographical, seasonal and gender variations in Tinospora cordifolia stem using HPLC-ESI-QTOF-MS. Phytochem Anal. 2017;28:277–288. doi: 10.1155/2012/18010310.1002/pca.2673. [DOI] [PubMed] [Google Scholar]
  2. Baker DA, Stevenson DW, Little DP. DNA barcode identification of black cohosh herbal dietary supplements. J AOAC Int. 2012;95:1023–1034. doi: 10.1155/2012/18010310.5740/jaoacint.11-261. [DOI] [PubMed] [Google Scholar]
  3. Cao Y, Fang S, Yin FuZ, Shang X, Yang W, Yang H. Chemical fingerprint and multicomponent quantitative analysis for the quality evaluation of Cyclocarya paliurus leaves by HPLC–Q–TOF–MS. Molecules. 2012;22:1927. doi: 10.3390/molecules22111927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. CBOL Plant working group a DNA barcode for land plants. Proc Natl Acad Sci USA. 2009;1061:2794–12797. doi: 10.1073/pnas.0905845106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chase MW, Cowan RS, Hollingsworth PM, van den Berg C, Madrinan S, Petersen G, Seberg O, Jorgsensen T, Cameron KM, Carine M, Pedersen N, Hedderson TAJ, Conrad F, Salazar GA, Richardson JE, Hollingsworth ML, Barraclough TG, Kelly L, Wilkinson M. A proposal for a standardised protocol to barcode all land plants. Taxon. 2007;56:295–299. doi: 10.1002/tax.562004. [DOI] [Google Scholar]
  6. Chaves OS, Gomes RA, Tomaz ACA, Fernandes MG, Mendes Junior LDG, Agra MF, Braga VA, Souza MFV. Secondary metabolites from Sida rhombifolia L. (Malvaceae) and the vaso-relaxant activity of cryptolepinone. Molecules. 2013;18:2769–2777. doi: 10.3390/molecules18032769. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. CISION PR News Wire (2020) India Ayurveda industry outlook 2019–2024—market size and growth forecasts, product launches, competitive landscape Report p. 161, Retrieved from https://www.prnewswire.com/newsreleases/india-ayurveda-industry-outlook-2019-2024—market-size–growthforecasts-product-launches-competitive-landscape-301007651.html. Accessed 24 Feb 2021
  8. Coghlan ML, Haile J, Houston J, Murray DC, White NE, Moolhuijzen P. Deep sequencing of plant and animal DNA contained within traditional Chinese medicines reveals legality issues and health safety concerns. PLoS Genet. 2012 doi: 10.1371/journal.pgen.1002657. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cortes N, Mora C, Munoz K, Díaz J, Serna R, Castro D, Osorio E. Microscopical descriptions and chemical analysis by HPTLC of Taraxacum officinale in comparison to Hypochaeris radicata a solution for mis-identification. Rev Bras Farmacogn. 2014;24:381–388. doi: 10.1016/j.bjp.2014.07.018. [DOI] [Google Scholar]
  10. De Boer HJ, Ichim MC, Newmaster SG. DNA barcoding and pharmacovigilance of herbal medicines. Drug Saf. 2015;38:611–620. doi: 10.1007/s40264-015-0306-8. [DOI] [PubMed] [Google Scholar]
  11. Dechbumroong P, Aumnouypol S, Denduangboripant J, Sukrong S. DNA barcoding of Aristolochia plants and development of species-specific multiplex PCR to aid HPTLC in ascertainment of Aristolochia herbal materials. PLoS ONE. 2018;13:e0202625. doi: 10.1371/journal.pone.0202625. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dev SA, Muralidharan EM, Sujanapal P, Balasundaran M. Identification of market adulterants in East Indian sandalwood using DNA barcoding. Ann Sci. 2013;71:517–522. doi: 10.1007/s13595-013-0354-0. [DOI] [Google Scholar]
  13. Dhalwal K, Shinde V. Optimization and validation of reverse phase HPLC and HPTLC method for simultaneous quantification of vasicine and vasicinone in Sida species. J Med Plant Res. 2010;4:1289–1296. [Google Scholar]
  14. Doebley J, Durbin M, Golenberg EM, Clegg MDP. Evolutionary analysis of the large subunit of carboxylase rbcL nucleotide sequence among the grasses Gramineae. Evolution. 1990;44:1097–1108. doi: 10.2307/2409569. [DOI] [PubMed] [Google Scholar]
  15. FAO and UNEP The state of the world’s forests 2020. Forests, biodiversity and people. Rome. 2020 doi: 10.4060/ca8642en. [DOI] [Google Scholar]
  16. Ganesan A, Marichamy K, Thagamariappan J. Development of medicinal plants sector in India—an empirical study. IJETMAS. 2016;4:82–89. [Google Scholar]
  17. Gautam RS, Mishra M, Prasad R (2020) Development and trade of medicinal and aromatic plants (MAPs): learnings from comparative analysis of MAPs export of India and China. Working Paper, Rajiv Gandhi Institute for Contemporary Studies (RGICS), New Delhi, p 59
  18. Goraya GS, Ved DK. Medicinal plants in India an assessment of their demand and supply. Dehradun: Ministry of AYUSH; 2017. [Google Scholar]
  19. Hall TA. BioEdit a user-friendly biological sequence alignment editor and analysis Program for Windows 95/98/NT. Nucleic Acids Sympo Ser. 1999;41:95–98. [Google Scholar]
  20. Hartvig I, Czako M, Kjær ED, Nielsen LR, Theilade I. The use of DNA barcoding in identification and conservation of rosewood (Dalbergia spp.) PLoS ONE. 2015 doi: 10.1371/journal.pone.0138231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. He T, Jiao L, Yu GJ, Jiang X, Yin Y. DNA barcoding authentication for the wood of eight endangered Dalbergia timber species using machine learning approaches. Holzforschung. 2018 doi: 10.1515/hf-2018-0076. [DOI] [Google Scholar]
  22. He T, Jiao L, Wiedenhoeft AC, Yin Y. Machine learning approaches outperform distance- and tree-based methods for DNA barcoding of Pterocarpus wood. Planta. 2019 doi: 10.1007/s00425-019-03116-3. [DOI] [PubMed] [Google Scholar]
  23. Heinrich M, Lardos A, Leonti M, Weckerle C, Willcox M, Applequist W, Stafford G. Best practice in research consensus statement on ethnopharmacological field studies–Con SEFS. J Ethnopharmacol. 2018;211:329–339. doi: 10.1016/j.jep.2017.08.015. [DOI] [PubMed] [Google Scholar]
  24. Hollingsworth ML, Andra AC, Forrest LL, Richardson J, Pennington RT, Long DG, Hollingsworth PM. Selecting barcoding loci for plants evaluation of seven candidate loci with species-level sampling in three divergent groups of land plants. Mol Ecol Resour. 2009;9:439–457. doi: 10.1111/j.1755-0998.2008.02439.x. [DOI] [PubMed] [Google Scholar]
  25. Ichim MC. The DNA-based authentication of commercial herbal products reveals their globally widespread adulteration. Front Pharmacol. 2019 doi: 10.3389/fphar.2019.0122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. SPSS Inc (2007) Released. SPSS for Windows, Version 160. SPSS Inc, Chicago
  27. Jadot I, Decleves A, Nortier J, Caron N. An integrated view of aristolochic acid nephropathy update of the literature. Int J Mol Sci. 2017 doi: 10.3390/ijms18020297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Joshi R, Satyal P, Setzer W. Himalayan aromatic medicinal plants a review of their ethnopharmacology, volatile phytochemistry and biological activities. Medicines. 2016 doi: 10.3390/medicines3010006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Joshi VK, Joshi A, Dhiman K. The ayurvedic pharmacopoeia of India, development and perspectives. J Ethnopharmacol. 2017 doi: 10.1016/j.jep.2016.07.030. [DOI] [PubMed] [Google Scholar]
  30. Kaur T, Bhat R, Vyas D. Effect of contrasting climates on antioxidant and bioactive constituents in five medicinal herbs in Western Himalayas. J Mt Sci. 2016;13:484–492. doi: 10.1007/s11629-014-3380-y. [DOI] [Google Scholar]
  31. Khatoon S, Srivastava M, Rawat A, Mehrotra S. HPTLC method for chemical standardization of Sida species and estimation of the alkaloid ephedrine. J Planar Chroma. 2005;18:364–367. doi: 10.1556/JPC.18.2005.5.5. [DOI] [Google Scholar]
  32. Kool A, de Boer HJ, Krüger A, Rydberg A, Abbad A, Björk L, Martin G. Molecular identification of commercialized medicinal plants in southern Morocco. PLoS ONE. 2012 doi: 10.1371/journal.pone.0039459. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kumar S, Stecher G, Tamura K. MEGA7 molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol Biol Evol. 2016;33:1870–1874. doi: 10.1093/molbev/msw054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kumar SJ, Gogna N, Newmaster SG, Venkatarangaiah K, Subramanyam R, Saroja SG, Ramanan US. DNA barcoding and NMR spectroscopy-based assessment of species adulteration in the raw herbal trade of Saraca asoca Roxb. Willd, an important medicinal plant. Int J Legal Med. 2016;130:1457–1470. doi: 10.1007/s00414-016-1436-y. [DOI] [PubMed] [Google Scholar]
  35. Lewis DD. Naïve Bayes at forty the independence assumption in information retrieval. Mach Learn. 1998;98:4–15. [Google Scholar]
  36. Little DP, Jeanson ML. DNA barcode authentication of Saw palmetto herbal dietary supplements. Sci Rep. 2013 doi: 10.1038/srep03518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Liu Y, Reich PB, Li G, Sun S. Shifting phenology and abundance under experimental warming alters trophic relationships and plant reproductive capacity. Ecology. 2011;92:1201–1207. doi: 10.1890/10-2060.1. [DOI] [PubMed] [Google Scholar]
  38. Meier R, Shiyang K, Vaidya G. DNA barcoding and taxonomy in Diptera a tale of high intraspecific variability and low identification success. Syst Biol. 2006;55:715–728. doi: 10.1080/10635150600969864. [DOI] [PubMed] [Google Scholar]
  39. Meyer CP, Paulay G. DNA barcoding error rates based on comprehensive sampling. PLoS ONE. 2005 doi: 10.1080/10635150600969864. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Michl J, Jennings HM, Kite GC, Ingrouille MJ, Simmonds MSJ, Heinrich M. Is aristolochic acid nephropathy a widespread problem in developing countries? A case study of Aristolochia indica L. in Bangladesh using an ethnobotanical-phytochemical approach. J Ethnopharmacol. 2013;149:235–244. doi: 10.1016/j.jep.2013.06.028. [DOI] [PubMed] [Google Scholar]
  41. Mishra P, Kumar A, Nagireddy A, Mani DN, Shukla AK, Tiwari R, Sundaresan V. DNA barcoding an efficient tool to overcome authentication challenges in the herbal market. J Plant Biotechnol. 2016 doi: 10.1111/pbi.12419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Mishra P, Kumar A, Nagireddy A, Shukla AK, Sundaresan V. Evaluation of single and multilocus DNA barcodes towards species delineation in complex tree genus Terminalia. PLoS ONE. 2017 doi: 10.1371/journal.pone.0182836. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. More RP, Mane RC, Purohit HC. MatK-QR classifier: a patterns based approach for plant species identification. BioData Min. 2016;9:39. doi: 10.1186/s13040-016-0120-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Moustafa MF, Hesham AEL, Quraishi MS, Alrumman SA. Variations in genetic and chemical constituents of Ziziphus spina-christi L. populations grown at various altitudinal zonation up to 2227m height. J Genet Eng Biotechnol. 2016;14:349–362. doi: 10.1016/j.jgeb.2016.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Newmaster SG, Grguric M, Shanmughanandhan M, Ramalingam S, Ragupathy S. DNA barcoding detects contamination and substitution in North American herbal products. BMC Med. 2013;11:222–235. doi: 10.1186/1741-7015-11-222. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
  46. Nithaniyal S, Parani M. Evaluation of chloroplast and nuclear DNA barcodes for species identification in Terminalia L. Biochem Syst Ecol. 2016;68:223–229. doi: 10.1016/j.bse.2016.08.001. [DOI] [Google Scholar]
  47. Ouarghidi A, Powell B, Martin GJ, HugoAbbad A. Species substitution in medicinal roots and possible implications for toxicity in Morocco. Econ Bot. 2013;66:370–382. doi: 10.1007/s12231-012-9215-2. [DOI] [Google Scholar]
  48. Palhares RM, Drummond MG, Brasil BS, Krettli AU, Oliveira GC, Brandao MG. The use of an integrated molecular, chemical, and biological based approach for promoting the better use and conservation of medicinal species a case study of Brazilian quinas. J Ethnopharmacol. 2014;155:815–822. doi: 10.1016/j.jep.2014.06.040. [DOI] [PubMed] [Google Scholar]
  49. Palhares RM, Gonçalves DM, Alves BDS, Cosenza GP, Lins MDG, Brandao OG. Medicinal plants recommended by the World Health Organization DNA barcode identification associated with chemical analyses guarantees their quality. PLoS ONE. 2015;10:e0127866. doi: 10.1371/journal.pone.0127866. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Parveen A, Adams JS, Raman V, Budel JM, Zhao BGNM, Ali Z, Khan IA. Comparative morpho-anatomical and HPTLC profiling of Tinospora species and dietary supplements. Planta Med. 2020 doi: 10.1055/a-1120-3711. [DOI] [PubMed] [Google Scholar]
  51. Paul C, Debnath A, Chakraborty K. Sex-specific variations in phytochemicals and antimicrobial potentiality of Dioscorea. Future J Pharm Sci. 2020 doi: 10.1186/s43094-020-00076-4. [DOI] [Google Scholar]
  52. Quinlan JR. Improved use of continuous attributes in C4.5. J Artif Intell Res. 1996;4:77–90. doi: 10.1613/jair.279. [DOI] [Google Scholar]
  53. Raclariu AC, Tebrencu CE, Ichim MC, Ciuperca OT, Brysting AK, de Boer H. What’s in the box? Authentication of Echinacea herbal products using DNA metabarcoding and HPTLC. Phytomedicine. 2018 doi: 10.1016/j.phymed.2018.03.058. [DOI] [PubMed] [Google Scholar]
  54. Roy S, Tyagi A, Shukla V, Kumar A, Singh UM, Chaudhary LB, Tuli R. Universal plant DNA barcode loci may not work in complex groups a case study with Indian Berberis Species. PLoS ONE. 2010 doi: 10.1371/journal.pone.0013674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Santhosh Kumar JU, Krishna V, Seethapathy GS, Ganesan R, Ravikanth G, Shanker RU. Assessment of adulteration in raw herbal trade of important medicinal plants of India using DNA barcoding. 3 Biotech. 2018 doi: 10.1007/s13205-018-1169-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Seethapathy GS, Tadesse M, Urumarudappa SKJV, Gunaga S, Vasudeva R, Malterud KE, Wangensteen H. Authentication of Garcinia fruits and food supplements using DNA barcoding and NMR spectroscopy. Sci Rep. 2018 doi: 10.1038/s41598-018-28635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Sgamma T, Masiero E, Mali P, Mahat M, Slater A. Sequence-specific detection of Aristolochia DNA – a simple test for contamination of herbal products. Front Plant Sci. 2018;91:828. doi: 10.3389/fpls.2018.01828. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Shahzad W, Asad S, Khan MA. Feature subset selection using association rule mining and JRip classifier. Int J Phys Sci. 2013;88:85–96. [Google Scholar]
  59. Sharma P, Sharma S, Yadav S, Srivastava A, Varma A, Luhana K, Shrivastava N. Categorization of Saraca asoca and Polyalthia longifolia, using chemical and genetic fingerprinting associated with multivariate statistical analysis. J Appl Res Med Aromat Plants. 2018;9:91–96. doi: 10.1016/j.jarmap.2018.03.004. [DOI] [Google Scholar]
  60. Shawky E, Abou EKM. Rapid discrimination of different Apiaceae species based on HPTLC fingerprints and targeted flavonoids determination using multivariate image analysis. Phytochem Anal. 2018 doi: 10.1002/pca.2749. [DOI] [PubMed] [Google Scholar]
  61. Simmler C, Graham JG, Chen SN, Pauli GF. Integrated analytical assets aid botanical authenticity and adulteration management. Fitoterapia. 2018;129:401–414. doi: 10.1016/j.fitote.2017.11.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Srirama R, Senthilkumar U, Sreejayan N, Ravikanth G, Gurumurthy BR, Shivanna MB, Uma Shaanker R. Assessing species admixtures in raw drug trade of Phyllanthus, a hepato-protective plant using molecular tools. J Ethnopharmacol. 2010;130:208–215. doi: 10.1016/j.jep.2010.04. [DOI] [PubMed] [Google Scholar]
  63. Srirama R, Santhosh Kumar JU, Seethapathy GS, Newmaster SG, Ragupathy S, Ganeshaiah KN, Ravikanth G. Species adulteration in the herbal trade: causes, consequences and mitigation. Drug Saf. 2017;40(8):651–661. doi: 10.1007/s40264-017-0527-0. [DOI] [PubMed] [Google Scholar]
  64. Stoeckle MY, Gamble CC, Kirpekar R, Young G, Ahmed S, Little DP. Commercial teas highlight plant DNA barcode identification successes and obstacle. Sci Rep. 2011 doi: 10.1038/srep00042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Suykens JAK, Vandewalle J. Least squares support vector machine classifiers. Neural Process Lett. 1999;9:293–300. doi: 10.1023/A:1018628609742. [DOI] [Google Scholar]
  66. Swetha VP, Parvathy VA, Sheeja TE, Sasikumar B. DNA barcoding for discriminating the economically important Cinnamomum verum from its adulterants. Food Biotechnol. 2014;28:183–194. doi: 10.1080/08905436.2014.931239. [DOI] [Google Scholar]
  67. Swetha VP, Parvathy VA, Sheeja TE, Sasikumar B. Authentication of Myristica fragrans Houtt. using DNA barcoding. Food Control. 2017;73:1010–1015. doi: 10.1016/j.foodcont.2016.10.004. [DOI] [Google Scholar]
  68. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6 molecular evolutionary genetics analysis version 6.0. Mol Biol Evol. 2013;30:2725–2729. doi: 10.1093/molbev/mst197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Techen N, Parveen I, Pan Z, Khan IA. DNA barcoding of medicinal plant material for identification. Curr Opin Biotechnol. 2014;25:103–110. doi: 10.1016/j.copbio.2013.09.010. [DOI] [PubMed] [Google Scholar]
  70. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. The CLUSTAL X windows interface flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997;25:4876–4882. doi: 10.1093/nar/25.24.4876. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Unnikrishnan R, Sumod M, Jayaraj R, Sujanapal P, Dev SA. The efficacy of machine learning algorithm for raw drug authentication in Coscinium fenestratum (Gaertn.) Colebr. employing a DNA barcode database. Physiol Mol Biol Plants. 2021;27(3):605–617. doi: 10.1007/s12298-021-00965-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Upton RT. Use of High-performance thin layer chromatography by the American herbal pharmacopoeia. J AOAC Int. 2010;93:1349–1354. doi: 10.1093/jaoac/93.5.1349. [DOI] [PubMed] [Google Scholar]
  73. Upton R, David B, Gafner S, Glasl S. Botanical ingredient identification and quality assessment strengths and limitations of analytical techniques. Phytochem Res. 2019 doi: 10.1007/s11101-019-09625-z. [DOI] [Google Scholar]
  74. Walker KM, AppleQuist WL. Adulteration of selected unprocessed botanicals in the US retail herbal trade. Econ Bot. 2012;66:321–327. doi: 10.1007/s12231-012-9211-6. [DOI] [Google Scholar]
  75. Wallace LJ, Boilard SMAL, Eagle SHC, Spall JL, Shokralla S, Hajibabaei M. DNA barcodes for everyday life routine authentication of natural health products. Food Res Int. 2012;49:446–452. doi: 10.1016/j.foodres.2012.07.048. [DOI] [Google Scholar]
  76. Weitschek E, Fiscon G, Felici G. Supervised DNA Barcodes species classification analysis, comparisons and results. Biodata Minining. 2014 doi: 10.1186/1756-0381-7-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. WHO . The World medicines situation—traditional medicines global situation, issues and challenges. Geneva: WHO Press; 2011. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary file1 (DOC 147 KB) (147KB, doc)
13205_2021_3001_MOESM2_ESM.jpg (87.8KB, jpg)

Supplementary file2 (JPG 88 KB) Agarose gel electrophoresis (a) (M- 100 bp ladder, matK, rbcL, psbA-trnH) (b) (M-100 bp ladder, ITS)

13205_2021_3001_MOESM3_ESM.jpg (186.1KB, jpg)

Supplementary file3 (JPG 186 KB) HPTLC banding pattern (a. S. alnifolia and its adulterants, b. D. gangeticum and its adulterants, c.T. arjuna and its adulterants, d. S. asoca and its adulterants)

Articles from 3 Biotech are provided here courtesy of Springer

RESOURCES