Biology subjects

Sathyaseelan, C.

Publications and source records attributed to Sathyaseelan, C..

3 recordsLinked to original sources

bioRxiv2022

G-quadruplex landscape and its regulation revealed by a new antibody capture method

Our understanding of DNA G-quadruplexes (G4s) from in vitro studies has been complemented by genome-wide G4 landscapes from cultured cells. Conventionally, the formation of G4s is accepted to depend on G-repeats such that they form tetrads. However, genome-wide G4s characterized through high-throughput sequencing suggest that these structures form at a large number of regions with no such canonical G4-forming signatures. Many G4-binding proteins have been described with no evidence for any protein that binds to and stabilizes G4s. It remains unknown what fraction of G4s formed in human cells are protein-bound. The G4-chromatin immunoprecipitation (G4-ChIP) method hitherto employed to describe G4 landscapes preferentially reports G4s that get crosslinked to proteins in their proximity. Our current understanding of the G4 landscape is biased against representation of G4s which escape crosslinking as they are not stabilized by protein-binding and presumably transient. We report a protocol that captures G4s from the cells efficiently without any bias as well as eliminates the detection of G4s formed artifactually on crosslinked sheared chromatin post-fixation. We discover that G4s form sparingly at SINEs. An application of this method shows that depletion of a repeat-binding protein CGGBP1 enhances net G4 capture at CGGBP1-dependent CTCF-binding sites and regions of sharp interstrand G/C-skew transitions. Thus, we present an improved method for G4 landscape determination and by applying it we show that sequence property-specific constraints of the nuclear environment mitigate G4 formation.

genomics↗

bioRxiv2020

Global variation in the SARS-CoV-2 proteome reveals the mutational hotspots in the drug and vaccine candidates

To accelerate the drug and vaccine development against the severe acute respiratory syndrome virus 2 (SARS-CoV-2), a comparative analysis of SARS-CoV-2 proteome has been performed in two phases by considering manually curated 31389 whole genome sequences from 84 countries. Among the 9 mutations that occur at a high significance (T85I-NPS2, L37F-NSP6, P323L-NSP12, D614G-spike, Q57H-ORF3a, G251V-ORF3a, L84S-ORF8, R203K-nucleocapsid and G204R-nucleocapsid), R203K-nucleocapsid and G204R-nucleocapsid are co-occurring (dependent) mutations and P323L-NSP12 and D614G-spike often appear simultaneously. Other notable variations that appear with a moderate to low significance are, M85-NSP1 deletion, D268-NSP2 deletion, 112 amino acids deletion in ORF8, a phenylalanine insertion amidst F34-F36 (NSP6) and several co-existing (dependent) substitution/deletion (I559V & P585S in NSP2, P504L & Y541C in NSP13, G82 & H83 deletions in NSP1 and K141, S142 & F143 deletions in NSP2) mutations. P323L-NSP12, D614G-spike, L37F-NSP6, L84S-ORF8 and the sequences deficient of the high significant mutations have led to 4 major SARS-CoV-2 clades. The top 5 countries bearing all the high significant and majority of the moderate significant mutations are: USA, England, Wales, Australia and Scotland. Further, the majority of the significant mutations have evolved in the first phase and have already transmitted around the globe indicating the positive selection pressure. Among the 26 SARS-CoV-2 proteins, nucleocapsid protein, ORF3a, ORF8, RNA dependent RNA polymerase and spike exhibit a higher heterogeneity compared with the rest of the proteins. However, NSP9, NSP10, NSP8, the envelope protein and NSP4 are highly resistant to mutations and can be exploited for drug/vaccine development.

bioinformatics↗

bioRxiv2020

Secondary structural characterization of the nucleic acids from circular dichroism spectra using extreme gradient boosting decision-tree algorithm

Nucleic acids exhibit a repertoire of conformational preference depending on the sequence and environment. Circular dichroism (CD) is an important and valuable tool for monitoring such secondary structural conformations of nucleic acids. Nonetheless, the CD spectral diversity associated with these structures poses a challenge in obtaining the quantitative information about the secondary structural content of a given CD spectrum. To this end, the competence of extreme gradient boosting decision-tree algorithm has been exploited here to predict the diverse secondary structures of nucleic acids. A curated library of 610 CD spectra corresponding to 16 different secondary structures of nucleic acids has been developed and used as a training dataset. For a test dataset of 242 CD spectra, the algorithm exhibited the prediction accuracy of 99%. For the sake of accessibility, the entire process is automated and implemented as a webserver, called CD-NuSS (CD to nucleic acids secondary structure) and is freely accessible at https://www.iith.ac.in/cdnuss/. The XGBoost algorithm presented here may also be extended to identify the hybrid nucleic acid topologies in future.

bioinformatics↗