Biology subjects

Swain, A. K.

Publications and source records attributed to Swain, A. K..

5 recordsLinked to original sources

bioRxiv2024

AP2/ERF transcription factors enriched in the drought-response transcriptome of the Thar desert tree Prosopis cineraria show higher copy number and greater DNA-binding affinity than orthologs in drought-sensitive species

We sequenced the drought-response transcriptome of the keystone tree species Prosopis cineraria from the Indian Thar desert to understand the key factors in its drought tolerance mechanism. We identified a network of genes activated in P. cineraria involved in osmotic stress response, phytohormone, calcium, and phosphorelay signal transduction. Of these, up-regulation of 54 APETALA2/Ethylene-Responsive Factor (AP2/ERF) transcription factor genes, validated by real-time PCR, suggests their key role in the drought tolerance of P. cineraria. We conducted a genome-wide study of the AP2/ERF superfamily in P. cineraria, classifying its 232 proteins into 15 clades and analyzing their protein structures, gene structure, and promoter organization. The P. cineraria genome contains more copies of AP2/ERF genes than drought-sensitive plants. Further, we identified sequence polymorphisms in AP2/ERF genes between Arabian and Indian cultivars of P. cineraria. We modeled the DNA-protein complex structures of AP2/ERFs from drought-tolerant and sensitive species using AlphaFold to compare their DNA binding ability. Though the DNA binding domain (DBD) is relatively conserved across species, the unstructured region of these proteins possesses different charge distributions, which might contribute differently to their DNA search and binding. Using all-atom molecular dynamics simulations, we teased out a higher number of specific DBD-DNA hydrogen bonds in P. cineraria, leading to a stronger DNA-binding affinity compared to drought-sensitive Arabidopsis thaliana. These results directly support copy number expansion of AP2/ERF transcription factors and the evolution of their structures for more efficient DNA search and binding as drought adaptation mechanisms in P. cineraria.

plant biology↗

bioRxiv2023

Deep learning-based method to identify disease-resistance proteins in Oryza sativa and relative species

Rice (Oryza sativa) is a significant agricultural crop consumed by more than half of the global population. Its demand is expected to increase due to rising consumption and a growing global population. Moreover, the rice plant is frequently exposed to disease-causing pathogens, such as bacteria, fungi, viruses, and nematodes. Thus, cultivating disease-resistant varieties is an efficient way of disease control compared to pesticide applications. However, the rice plant has a well-defined defense system to prevent the onset of disease, including Pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI) and effector-triggered immunity (ETI). The defense system is controlled by various disease-resistance proteins, such as resistance (R) proteins and pathogen recognition receptors (PRRs). Therefore, the identification of disease-resistance proteins not only reduces the amount of pesticides used in rice fields but also increases their yield. Though some resistant proteins have been characterized, their rapid identification, precise diagnosis, and appropriate management are still lacking. However, few methods based on sequence-similarity and de novo prediction, such as Machine Learning (ML), usually have low prediction power. In this study, we built a state-of-the-art classifier based on Deep Learning (DL) for the early detection of disease-resistance proteins in rice and related species. We compared the DL-based Multi-layer Perceptron (MLP) model with the five well-established ML-based methods using a protein dataset of rice and its related species. The DL-based MLP model outperformed all of the five classifiers on 10-fold cross-validation. The accuracy, Area Under Receiving Operating Characteristic (ROC) curve (AUC), F1-score, precision, and recall were superior in the DL-based MLP model. In conclusion, the MLP model is an effective DL model for predicting disease-resistance proteins with high scores in performance metrics. This study will provide insight to the breeders in developing disease-resistant rice varieties and assist in transforming traditional rice farming practices into a new age of smart rice farming.

bioinformatics↗

bioRxiv2023

SpatialPrompt: spatially aware scalable and accurate tool for spot deconvolution and clustering in spatial transcriptomics

Spatial transcriptomics has advanced our understanding of tissue biology by enabling sequencing while preserving spatial coordinates. In sequencing-based spatial technologies, each measured spot typically consists of multiple cells. Deconvolution algorithms are required to decipher the cell-type distribution at each spot. Existing spot deconvolution algorithms for spatial transcriptomics often neglect spatial coordinates and lack scalability as datasets get larger. We introduce SpatialPrompt, a spatially aware and scalable method for spot deconvolution as well as domain identification for spatial transcriptomics. Our method integrates gene expression, spatial location, and single-cell RNA sequencing (scRNA-seq) reference data to infer cell-type proportions of spatial spots accurately. At the core, SpatialPrompt uses non-negative ridge regression and an iterative approach inspired by graph neural network (GNN) to capture the local microenvironment information in the spatial data. Quantitative assessments on the human prefrontal cortex dataset demonstrated the superior performance of our tool for spot deconvolution and domain identification. Additionally, SpatialPrompt accurately decipher the spatial niches of the mouse cortex and the hippocampus regions that are generated from different protocols. Furthermore, consistent spot deconvolution prediction from multiple references on the mouse kidney spatial dataset showed the impressive robustness of the tool. In response to this, SpatialPromptDB database is developed to provide compatible scRNA-seq references with cell-type annotations for seamless integration. In terms of scalability, SpatialPrompt is the only method performing spot deconvolution and clustering in less than 2 minutes for large spatial datasets with 50,000 spots. SpatialPrompt tool along with the SpatialPromptDB database are publicly available as open source software for large-scale spatial transcriptomics analysis.

bioinformatics↗

bioRxiv2022

Comparison of Dimensionality Reduction and Clustering Methods for Single-Cell Transcriptomics Data

Dimensionality reduction (DR) methods are applied to extract relevant features from inherently high dimensional and noisy single-cell RNA sequencing (scRNA-seq) data. Choice of DR method could influence the performance of clustering algorithm and subsequent analysis outcomes. We performed a benchmarking study of seven popular DR methods and four clustering algorithms widely used for scRNA-seq datasets. For this purpose, we used three publicly available real scRNA-seq datasets. The performance was evaluated using two clustering metrics viz. adjusted random index (ARI) and normalized mutual index (NMI). We also compared our results with a similar study published by Xiang and colleagues. Overall, we observed higher ARI and NMI scores for DR methods when compared with Xiangs study. We also noticed several differences between our and Xiangs study. Noteworthy, three methods, namely, Independent Component Analysis (ICA), t-Distributed Stochastic Neighbor Embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP) performed consistently well across three datasets. Linear method ICA was best performer on Segerstolpe dataset, while nonlinear methods UMAP and t-SNE best performed on Deng and Chu datasets, respectively. Neural network-based methods Variational Autoencoder (VAE) and Deep Count Autoencoder (DCA) could not perform well probably due to their sensitivity to hyperparameters and overfitting. Among clustering methods, Gaussian Mixture Models (GMMs) performed consistently well across datasets. This might be because GMMs are the universal approximators of posterior probability densities. We conclude that performance of different DR methods is more dataset dependent and for various scRNA-seq datasets different algorithms are more suited and there is no one-fit-all method.

bioinformatics↗

bioRxiv2022

Single-Cell Transcriptome Analysis Identifies Novel Biomarkers Involved in Major Liver Cancer Subtypes

Liver cancers including hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC) are leading cause of death worldwide. Single-cell transcriptomics studies have vast potential in advancing our understanding of cancers by defining the cellular composition of different solid tumor types. We peformed an integrated analysis using single-cell RNA sequencing (scRNA-seq) data from cancerous and healthy liver tissues in order to identify the molecular progression and intercellular heterogeneity across cell types in the liver cancer. Moreover, we performed a subtype specific analyses, separately for HCC and ICC, to identify any molecular drivers uniquely associated with these liver cancers. The scRNA-seq dataset comprising 5 healthy controls and 19 liver cancer patients were collected from Human Cell Atlas and Gene Expression Omnibus (GEO), respectively. Our analyses confirmed upregulation of four previously known malignant cell marker genes, namely, EPCAM, KRT19, KRT7 and S100P in the cancerous liver cells. Of these, KRT7 gene has been reported to be associated with ovarian cancer in the past studies. Noteworthy, four marker genes specific to the G1/S (MCM5 and PCNA) and G2/M phases (HMGB2 and CKS2) of the cell cycle were upregulated in the cancerous liver cells. This indicates that these four marker genes are actively dividing in these two phases in cancerous cells as compared to normal liver cells. Our differential expression analysis identified 2 upregulated genes (ATF3 and S100A11) and 2 downregulated genes (FCN3 and FGB) in the liver cancer. Our subtype based differential expression analysis identified 4 genes (HSPA6, LMNA, ATP1B1 and DCXR) specific to HCC and 3 genes (HSPB1, APOC3 and APOA1) specific to ICC. CD4+ T-cell, Hepatocyte, neutrophil, mesenchymal cells and liver bud hepatic cells are the predominant cell-types in liver cells. Our scRNA-seq study revealed the mesenchymal cells as potential malignant cell types in liver cancers. Our work suggests future research on developing liver cancer subtypes therapies could target these cell types and associated molecular markers.

genomics↗