Biology subjects

Madhu, P.

Publications and source records attributed to Madhu, P..

2 recordsLinked to original sources

bioRxiv2021

Sub-stoichiometric Hsp104 regulates the genesis and persistence of self-replicable amyloid seeds of a yeast prion protein

The prion-like self-perpetuating conformational conversion is involved in both transmissible neurodegenerative diseases and non-Mendelian inheritance traits. The transmissibility of amyloid-like aggregates is dependent on the stoichiometry of chaperones such as heat shock proteins. To provide the mechanistic underpinning of the generation and persistence of prefibrillar amyloid seeds that are critical for the prion-like propagation, we studied the effect of Hsp104 disaggregase on the assembly mechanism of a yeast prion determinant of Saccharomyces cerevisiae Sup35. At low sub-stoichiometric concentrations, Hsp104 exhibits a dual role and considerably accelerates the formation of seeding-competent prefibrillar amyloids by shortening the lag phase but also prolongs their persistence by introducing unusual kinetic halts and delaying their conversion into matured fibers. Hsp104-mediated amyloid species comprise a more ordered packing and display an enhanced autocatalytic self-templating ability compare to amyloids formed without Hsp104. Our findings underscore the key functional and pathological roles of sub-stoichiometric chaperones in prion-like propagation.

biophysics↗

bioRxiv2021

Assigning Secondary Structure in Proteins using AI

Knowledge about protein structure assignment enriches the structural and functional understanding of proteins. Accurate and reliable structure assignment data is crucial for secondary structure prediction systems. Since the 80s various methods based on hydrogen bond analysis and atomic coordinate geometry, followed by Machine Learning, have been employed in protein structure assignment. However, the assignment process becomes challenging when missing atoms are present in protein files. Our model develops a multi-class classifier program named DLFSA for assigning protein Secondary Structure Elements(SSE) using Convolutional Neural Networks(CNN). A fast and efficient GPU based parallel procedure extracts fragments from protein files. The model implemented in this work is trained with a subset of protein fragments and achieves 88.1% and 82.5% train and test accuracy, respectively. Our model uses only C coordinates for secondary structure assignments. The model is successfully tested on a few full-length proteins also. Results from the fragment-based studies demonstrate the feasibility of applying deep learning solutions for structure assignment problems.

bioinformatics↗