Quality control of jati belanda leaves (Guazuma ulmifolia) Using portable photometer and pattern recognition technic

Abstract

Jati belanda (Guazuma ulmifolia) is one of traditional medicinal plants. Quality control of medicinal plants is necessary to control the quality and safety of medicines. In this study, various qualities of jati belanda leaves was evaluated using a portable photometer combined with chemometric pattern recognition techniques PCA and PLSDA. Total flavonoids leaves at age 1, 2, and 3 months showed various quality of active compounds. PCA analysis the leaves with 7 light sources using the first 2 principle component (PC) was able to describe 94% of total variant in the data (PC1 = 86%, PC2 = 8%). Combination of 3 LED (green, blue, and blue-purple LEDs) provided the same pattern in the PCA score plot analysis using the 7 LED lights. Analysis PLSDA with 7 LED lights resulted 3 models, namely models PLSDA at age 1 month (R2 = 0.9674), 2 month (R2 = 0.8781), and 3 months (R2 = 0.9430). The model has successfully predicted samples of jati belanda leaves at age 1 month. PLSDA models with 7 LED lights was still better than PLSDA model with 3 LED light from the standpoint of from modeling indicator goodness (R2, RMSEC, RMSEP). PLSDA models for jati belanda leaves with 3 light sources (green, blue, and blue-purple LED) is sensitive enough to classify and predict the sample being tested.

Jati belanda (Guazuma Ulmifolia) merupakan tanaman obat tradisional. Kendali mutu tanaman diperlukan untuk mengontrol kualitas dan keamanan obat. Pada penelitian ini keragaman mutu daun jati belanda dievaluasi menggunakan fotometer jinjing yang dikombinasikan dengan teknik pengenalan pola kemometrik PCA dan PLSDA. Analisis total flavonoid menunjukkan daun jati belanda berumur 1, 2, dan 3 bulan memiliki keragaman mutu senyawa aktif. Analsis PCA daun jati belanda dengan 7 sumber lampu menggunakan 2 PC pertama mampu menjelaskan 94% dari keragaman total (PC1 86% dan PC2 8%). Kombinasi 3 lampu (LED hijau, LED biru, dan LED biru-ungu) memberikan kesamaan pola plot skor PCA seperti pada analisis menggunakan 7 lampu LED. Analisis PLSDA dengan 7 lampu LED menghasilkan 3 model, yaitu model PLSDA umur 1 bulan (R2 = 0.9674), 2 bulan (R2 = 0.8781), dan 3 bulan (R2 = 0.9430). Model tersebut telah berhasil memprediksi sampel daun berumur 1 bulan. Model PLSDA 7 lampu masih lebih baik daripada model PLSDA dengan 3 lampu jika dilihat dari indikator kebaikan pemodelan (nilai R2, RMSEC, RMSEP) yang didapatkan. Model PLSDA daun dengan 3 sumber lampu (LED hijau, LED biru, LED biru-ungu) masih cukup sensitif untuk mengklasifikasikan dan memprediksi sampel yang diujikan.