ABSTRACT
Single-cell RNA sequencing (scRNA-seq) is a revolutionary methodology that helps to analyze transcriptome or genome information from a single cell. However, high dimensionality and sparsity in data due to dropout events pose computational challenges for existing state-of-the-art scRNA-seq clustering methods. Learning efficient representations becomes even more challenging due to the presence of noise in scRNA-seq data. To overcome the effect of noise and learn effective representations, this paper proposes sc-INDC (Single-Cell Information Maximized Noise-Invariant Deep Clustering), a deep neural network that facilitates learning of informative and noise-invariant representations of scRNA-seq data. Furthermore, the time complexity of the proposed sc-INDC is significantly lower compared to state-of-the-art scRNA-seq clustering methods. Extensive experimentation on fourteen publicly available scRNA-seq datasets illustrates the efficacy of the proposed model. Additionally, visualizations of t-SNE plots and several ablation studies are also conducted to provide insights into the improved representation ability of sc-INDC. Code of the proposed sc-INDC will be available at: https://github.com/arnabkmondal/sc-INDC.
Subject(s)
Gene Expression Profiling , Single-Cell Analysis , Gene Expression Profiling/methods , Sequence Analysis, RNA/methods , Base Sequence , Single-Cell Analysis/methods , Cluster Analysis , AlgorithmsABSTRACT
The COrona VIrus Disease 2019 (COVID-19) pandemic is an ongoing global pandemic that has claimed millions of lives till date. Detecting COVID-19 and isolating affected patients at an early stage is crucial to contain its rapid spread. Although accurate, the primary viral test 'Reverse Transcription Polymerase Chain Reaction' (RT-PCR) for COVID-19 diagnosis has an elaborate test kit, and the turnaround time is high. This has motivated the research community to develop CXR based automated COVID-19 diagnostic methodologies. However, COVID-19 being a novel disease, there is no annotated large-scale CXR dataset for this particular disease. To address the issue of limited data, we propose to exploit a large-scale CXR dataset collected in the pre-COVID era and train a deep neural network in a self-supervised fashion to extract CXR specific features. Further, we compute attention maps between the global and the local features of the backbone convolutional network while finetuning using a limited COVID-19 CXR dataset. We empirically demonstrate the effectiveness of the proposed method. We provide a thorough ablation study to understand the effect of each proposed component. Finally, we provide visualizations highlighting the critical patches instrumental to the predictive decision made by our model. These saliency maps are not only a stepping stone towards explainable AI but also aids radiologists in localizing the infected area.
ABSTRACT
The present study was intended to determine the possible influences of direct field application of choline chloride into pond water in addition to farm-made-aqua-feed under a semi-intensive culture system on the growth and biochemical parameters of two Indian major carps (IMCs), Catla catla (Catla) and Labeo rohita (Rahu), and two air-breathing species, Clarias batrachus (Magur) and Anabas testudineus (Koi), cultured in a ratio of 2:5:1:1 (Catla/Rahu/Magur/Koi) in three experimental ponds for a period of 90 days during the breeding season (June to August). Results were compared with control (C: fed only with farm-made-aqua-feed) and treatment (T: P1 and P2: farm-made-aqua-feed plus choline chloride into pond water directly at the rate of 350 g bigha-1 fortnightly or 350 g per 1600 square meter fortnightly). A significant increasing trend was observed in the growth parameters including total length-final (TLF), standard length-final (SLF), mean weight-final (MWF), % gain of mean total length (MTL), % gain of mean standard length (MSL), % weight gain (WG), specific growth rate (SGR) % per day, and survivability %. However, a reverse pattern was noticed in the food conversion ratio (FCR) both in IMCs and air-breathing fish species under choline supplementation. Serum biochemical responses, e.g., total protein (PRO), lactate dehydrogenase (LDH), glucose (GLU), and calcium (Ca) showed significant enhancement, and alkaline phosphatase (ALP), alanine amino transaminase (ALT), aspartate amino transaminase (AST), cholesterol (CHOLES), and triglycerides (Trig) showed gradual significant reduction during the breeding season under choline exposure. Treated fishes showed prevention from liver dysfunction and fatty liver formation, and increased body crude protein content. Results indicated favorable growth and yield, which may benefit fish farmers during their culture practices, and the output fish species under choline supplementation resulted in quality food-fish for human consumption.
ABSTRACT
Clustering single-cell RNA sequence (scRNA-seq) data poses statistical and computational challenges due to their high-dimensionality and data-sparsity, also known as 'dropout' events. Recently, Regularized Auto-Encoder (RAE) based deep neural network models have achieved remarkable success in learning robust low-dimensional representations. The basic idea in RAEs is to learn a non-linear mapping from the high-dimensional data space to a low-dimensional latent space and vice-versa, simultaneously imposing a distributional prior on the latent space, which brings in a regularization effect. This paper argues that RAEs suffer from the infamous problem of bias-variance trade-off in their naive formulation. While a simple AE wita latent regularization results in data over-fitting, a very strong prior leads to under-representation and thus bad clustering. To address the above issues, we propose a modified RAE framework (called the scRAE) for effective clustering of the single-cell RNA sequencing data. scRAE consists of deterministic AE with a flexibly learnable prior generator network, which is jointly trained with the AE. This facilitates scRAE to trade-off better between the bias and variance in the latent space. We demonstrate the efficacy of the proposed method through extensive experimentation on several real-world single-cell Gene expression datasets. The code for our work is available at https://github.com/arnabkmondal/scRAE.
Subject(s)
Gene Expression Profiling , Single-Cell Analysis , Cluster Analysis , Gene Expression , Gene Expression Profiling/methods , Sequence Analysis, RNA/methods , Single-Cell Analysis/methodsABSTRACT
Objective: Since its outbreak, the rapid spread of COrona VIrus Disease 2019 (COVID-19) across the globe has pushed the health care system in many countries to the verge of collapse. Therefore, it is imperative to correctly identify COVID-19 positive patients and isolate them as soon as possible to contain the spread of the disease and reduce the ongoing burden on the healthcare system. The primary COVID-19 screening test, RT-PCR although accurate and reliable, has a long turn-around time. In the recent past, several researchers have demonstrated the use of Deep Learning (DL) methods on chest radiography (such as X-ray and CT) for COVID-19 detection. However, existing CNN based DL methods fail to capture the global context due to their inherent image-specific inductive bias. Methods: Motivated by this, in this work, we propose the use of vision transformers (instead of convolutional networks) for COVID-19 screening using the X-ray and CT images. We employ a multi-stage transfer learning technique to address the issue of data scarcity. Furthermore, we show that the features learned by our transformer networks are explainable. Results: We demonstrate that our method not only quantitatively outperforms the recent benchmarks but also focuses on meaningful regions in the images for detection (as confirmed by Radiologists), aiding not only in accurate diagnosis of COVID-19 but also in localization of the infected area. The code for our implementation can be found here - https://github.com/arnabkmondal/xViTCOS. Conclusion: The proposed method will help in timely identification of COVID-19 and efficient utilization of limited resources.
Subject(s)
COVID-19 , Deep Learning , COVID-19/diagnostic imaging , Humans , Radiography, Thoracic , SARS-CoV-2 , X-RaysABSTRACT
Polyaromatic compounds are the major, widespread contaminants in the aquatic environment. However, the adverse impacts of these compounds on blood pathophysiology (hematological profiling and serum biochemical responses) are poorly understood. As a consequence, this study was intended to evaluate the toxic effects of naphthalene, one of the polycyclic aromatic hydrocarbons, on the blood pathophysiology of Anabas testudineus using multiple end-point biomarker approach. A. testudineus was exposed to short-term (1 and 5 d) and long-term (10, 15, and 21 d) naphthalene concentrations, that is, T1 (0.71 mg/L indicates 25% of LC50) and T2 (1.42 mg/L indicates 50% of LC50 value). The results disclosed significant decrease in red blood cells, hemoglobin (Hb), packed cell volume, and platelet levels, while other blood parameters, namely, white blood cells, percent lymphocyte, mean cell volume, mean corpuscular Hb, and mean corpuscular Hb concentration showed enhanced levels under naphthalene intoxication. Results were more detrimental under T2 concentration. Cholesterol, glucose, calcium, high-density lipoprotein, and low-density lipoprotein levels gradually increased throughout the different exposure periods under T1 and T2 concentrations, while the triglyceride level gradually decreased during exposure periods. Finally, integrated biomarker responses (IBR) analysis indicated that serum biochemical parameters are more powerful than hematological parameters for determining the naphthalene-induced fish health status. Additionally, the IBR study clearly identified that long-term (>5 d) exposure was more harmful than short-term (<5 d) naphthalene exposure. So, these responses may be derived as biomarkers for monitoring naphthalene pollution in an aquatic ecosystem.
ABSTRACT
Anabas testudineus (Bloch) was exposed to 0.71â¯mg/L and 1.42â¯mg/L (25 and 50% of LC50 value respectively) naphthalene, a polycyclic aromatic hydrocarbon (PAH), for 21 days. Blood biochemical parameters and erythrocytic morphological alterations were assessed to describe the naphthalene toxicity. Biochemical analysis showed a significant increase in glutamic pyruvic transaminase, GPT (576.7 ± 11.79 and 608.9 ± 12.08 U/L, respectively) and alkaline phosphatase, ALP (12.9 ± 0.69 and 13.4⯱â¯0.64 U/L, respectively) activities under two doses compared with control. Protein and albumin (ALB) content in blood decreased significantly, in comparison with control value in the tune of 22.67⯱â¯1.04 and 23.97⯱â¯1.24â¯g/dl, respectively and 10.7⯱â¯0.79 and 11.1⯱â¯0.67â¯g/dl, respectively. Erythrocytes showed varied symptomatic morphological changes under naphthalene exposure, which included severe denaturation, swelling in cells, appearance of sickle and tear drop cells, and cellular vacuolation. In particularly, the changes were more prominent under higher naphthalene exposure. Following the results, it has been able to establish that GPT, ALP, protein and ALB, and the morphological manifestations of erythrocytes would be good tools of biomarker in monitoring toxicological paradigm, especially to naphthalene exposure in aquatic bodies.
Subject(s)
Erythrocytes/drug effects , Fishes/blood , Naphthalenes/toxicity , Water Pollutants, Chemical/toxicity , Alanine Transaminase/blood , Alkaline Phosphatase/blood , Animals , Erythrocytes/pathology , Fish Proteins/blood , Serum Albumin/analysisABSTRACT
Present study highlighted the ultramicroscopic (SEM) alterations of the skin, eye, barbel, and fins of spawn of an air-breathing teleost (Clarias batrachus, Linn. 1758) induced by UV-B radiation (280-320 nm) at a dose (@4.07 × 10-20J/photon/m2) under the time-frame of 5, 10 and 15 min/d in the laboratory condition for the periods of 5 and 10 days. Limnological parameters revealed no significant changes throughout the period of experimentation which were measured by PCS Testr 35 Multi-Parameter. Morphometric analysis revealed that during the extended exposure period of 10 days the spawn size and weight were reduced as analysed through Specific Growth Rate (SGR). SGR values in terms of weight for 5 and 10 days under 3 time-frames were 17.12%, 12.52%, 11.46% and 9.09%, 6.43%, 6.09% respectively, which revealed a declined trend along with the exposure days. In the skin of C. batrachus, the compact regular orientation of the stratified epithelial cells and mucous cells became distorted and the microridges and double-ridged structures showed destruction and fragmentations. The body striations and microfolds became shrinked and swollen and finally degenerated to form a mass. The distribution of mucous cells throughout the epidermis was disorganised and releasing secretory contents on the surface through small pores. Appearance of huge quantity of biogenic semi-hexagonal plate like crystals (guanine platelets) on the skin surface of the body was the most significant observations during UV-B radiation. In the developmental phases the eyeball showed shrinkage loosing normal regular concave structure and to become a dome-shaped one. The supportive connective infoldings became loosened. The choroid coat displayed deformities and the iris deformed the pupil. The fibroblast on the epithelium and melanocytes depicted dispersed arrangement. The pairs of ventral barbels near the mouth depicted the presence of taste buds that became severely damaged exposing the sensory as well as neuroepithelial cells. Compact regular arrangement of the SECs was completely destroyed leaving long and deep channels inbetween them; the disintegrated concentric MRs also showed a mass.