Ravi Sankar

Dermatan sulfate (DS) is a remarkably versatile glycosaminoglycan that plays critical roles across a wide array of biological processes. Its unique structure, characterized by repeating disaccharide units of N-acetyl-D-galactosamine (GalNAc) and Iduronic acid (IdoA) with variable sulfation patterns, enables it to interact with numerous biomolecules. These interactions mediate diverse functions, including the organization of the extracellular matrix, promotion of wound healing, and modulation of cancer progression. Despite its broad biological relevance, deciphering DS function remains challenging due to its pronounced structural complexity and heterogeneity. Variations in chain length, disaccharide composition, and sulfation patterns make it difficult to fully characterize DS’s intricate structure–function relationships. In this review, recent developments in biosynthesis, preparation, and applications of DS are summarized.

This article presents the design and the numerical analysis of a smart label-free Surface Plasmon Resonance (SPR) sensor to detect the concentration of haemoglobin in blood and the concentration of glucose in the urine samples. The suggested sensor uses a thin film of silver (Ag) on the prism’s surface to excite the surface plasmons. The finite element method (FEM) was used to do numerical simulations to optimize the layer thickness and to analyse the sensor’s performance in terms of the sensitivity and the Figure of merit (FOM). The results of the simulation showed that there is a linear correlation between resonance wavelength shift and change in analyte refractive index. The optimised design obtained a sensitivity of 288.29 °/RIU, QF of 780.80 $$\:{RIU}^{-1}$$, SNR of 15.62, FoM of 492.51 $$\:{RIU}^{-1}$$ and CSF of 539.20. The label-free methodology involves no chemical tagging and thus allows biosensing that is quick, real-time and economical. The suggested SPR sensor has great possibilities to be implemented in the non-invasive biomedical applications, diagnostics and point-of-care monitoring. In addition, machine learning models were employed to predict sensor sensitivity based on structural and optical parameters, demonstrating the strong capability of data-driven approaches for rapid performance estimation and design optimization.