Predictive modeling for ubiquitin proteins through advanced machine learning technique.

Ubiquitination is an essential post-translational modification mechanism involving the ubiquitin protein's bonding to a substrate protein. It is crucial in a variety of physiological activities including cell survival and differentiation, and innate and adaptive immunity. Any alteration in the ubiquitin system leads to the development of various human diseases. Numerous researches show the highly reversibility and dynamic of ubiquitin system, making the experimental identification quite difficult. To solve this issue, this article develops a model using a machine learning approach, tending to improve the ubiquitin protein prediction precisely. We deeply investigate the ubiquitination data that is proceed through different features extraction methods, followed by the classification. The evaluation and assessment are conducted considering Jackknife tests and 10-fold cross-validation. The proposed method demonstrated the remarkable performance in terms of 100 %, 99.88 %, and 99.84 % accuracy on Dataset-I, Dataset-II, and Dataset-III, respectively. Using Jackknife test, the method achieves 100 %, 99.91 %, and 99.99 % for Dataset-I, Dataset-II and Dataset-III, respectively. This analysis concludes that the proposed method outperformed the state-of-the-arts to identify the ubiquitination sites and helpful in the development of current clinical therapies. The source code and datasets will be made available at Github.

Spatial distribution of drinking and irrigation water quality in different climatic zones of Baluchistan, Pakistan.

Baluchistan's water profile was developed by dividing it into seven zones (Northern Highlands-NH, Southern Highlands-SH, Quetta Valley-QV, Desert-D, Sibbi Plains-SP, Coastal Lasbella-CL, Coastal Gwadar-CG) based on geography, water availability, and climate of the area. A total of 106 water samples were collected from karaiz, spring water, and tube wells. Spatial distribution of EC, TDS, TH, SO42-, Cl-, Na+, and K+ showed an increasing trend in concentration from the highlands towards the desert and coastal zones. For anion, HCO3- is predominant in NH, SH, and QV, Cl- in D, CL, and CG and only SO42- in SP, whereas the cationic trend in overall zones is Na+>Ca2+>Mg2+>K+. In the NH, SH, QV, and SP zones, the physicochemical parameters met the drinking water quality guidelines; however, D, CL, and CG exceeded in almost all quality parameters. Furthermore, the drinking water quality index (WQI) shows excellent to good water quality in NH, SH, QV, and D zones, while CL and CG fall in poor to unsuitable water classes. In terms of hydrogeochemical facies, maximum water samples from NH fall in Ca-Mg-HCO3, and SH, QV, and SP in Ca-Mg-Cl type, where major ion chemistry is controlled by rock-weathering, while D, CL, and CG fall in the NaCl type, where evaporation is dominant. Similarly, irrigation water quality parameters (EC, SAR, RSC, Na%, MH%, PI, SSP, and KR) reveal that NH, SH, QV, and SP have suitable water for irrigation, and D, CL, and CG require proper treatment. Additionally, USSL and Wilcox's diagrams indicated that NH, SH, QV, and SP have "excellent to permissible"; however, D, CL, and CG have "permissible to unsuitable" class water, requiring special management practices. Consequently, appropriate control measures and targeted water purification programmes should be implemented to protect the public health and sustainability of water resources in Baluchistan.

Adsorption and immobilization performance of pine-cone pristine and engineered biochars for antimony in aqueous solution and military shooting range soil: An integrated novel approach.

Antimony (Sb-V), a carcinogenic metalloid, is becoming prevalent in water and soil due to anthropogenic activities. Biochar could be an effective remedy for Sb(V)-contaminated water and soil. In this study, we used pristine and engineered pinecone-derived biochar as an innovative approach for treating Sb(V)-contaminated water and shooting range soil. Biochar was produced from pine-cone waste (pristine biochar) and enriched with Fe and Al salts via saturation (engineered biochar). Adsorption tests in water revealed that iron-modified biochar showed higher adsorption capacity (8.68 mg g-1) than that of the pristine biochar (2.49 mg g-1) and aluminum-modified biochar (3.40 mg g-1). Isotherm and kinetic modeling of the adsorption data suggested that the adsorption process varied from monolayer to multilayer, with chemisorption as the dominant interaction mechanism between Sb(V) and the biochars. The post-adsorption study of iron-modified biochar by Fourier Transform Infrared (FTIR) and X-ray diffraction (XRD) further supported the chemical bonding and outer-sphere complexation of Sb(V) with Fe, N-H, O-H, C-O and CC components. The pristine and iron-modified biochars also successfully immobilized Sb(V) in a shooting range soil, more so in the latter. Subsequent sequential extractions and post-analysis by scanning electron microscopy (SEM), energy dispersive X-ray analysis (EDX), and elemental dot mapping revealed that Sb in the treated soil transformed to a more stable form. It was concluded that iron-modified biochar could act as an efficient material for the adsorption and immobilization of Sb(V) in water and soil, respectively.

Violence Detection Using Spatiotemporal Features with 3D Convolutional Neural Network.

The worldwide utilization of surveillance cameras in smart cities has enabled researchers to analyze a gigantic volume of data to ensure automatic monitoring. An enhanced security system in smart cities, schools, hospitals, and other surveillance domains is mandatory for the detection of violent or abnormal activities to avoid any casualties which could cause social, economic, and ecological damages. Automatic detection of violence for quick actions is very significant and can efficiently assist the concerned departments. In this paper, we propose a triple-staged end-to-end deep learning violence detection framework. First, persons are detected in the surveillance video stream using a light-weight convolutional neural network (CNN) model to reduce and overcome the voluminous processing of useless frames. Second, a sequence of 16 frames with detected persons is passed to 3D CNN, where the spatiotemporal features of these sequences are extracted and fed to the Softmax classifier. Furthermore, we optimized the 3D CNN model using an open visual inference and neural networks optimization toolkit developed by Intel, which converts the trained model into intermediate representation and adjusts it for optimal execution at the end platform for the final prediction of violent activity. After detection of a violent activity, an alert is transmitted to the nearest police station or security department to take prompt preventive actions. We found that our proposed method outperforms the existing state-of-the-art methods for different benchmark datasets.