In recent years, the rapid growth of biological data has increased interest in using bioinformatics to analyze and interpret this data. Proteomics, which studies the structure, function, and ...interactions of proteins, is a crucial area of bioinformatics. Using natural language processing (NLP) techniques in proteomics is an emerging field that combines machine learning and text mining to analyze biological data. Recently, transformer‐based NLP models have gained significant attention for their ability to process variable‐length input sequences in parallel, using self‐attention mechanisms to capture long‐range dependencies. In this review paper, we discuss the recent advancements in transformer‐based NLP models in proteome bioinformatics and examine their advantages, limitations, and potential applications to improve the accuracy and efficiency of various tasks. Additionally, we highlight the challenges and future directions of using these models in proteome bioinformatics research. Overall, this review provides valuable insights into the potential of transformer‐based NLP models to revolutionize proteome bioinformatics.
Protein function prediction is one of the well-known problems in proteome research, attracting the attention of numerous researchers. However, the implementation of deep neural networks, which helps ...to increase the protein function prediction, still poses a big challenge. This study proposes a deep learning approach namely Fertility-GRU that incorporates gated recurrent units and position-specific scoring matrix profiles to predict the function of fertility-related protein, which is a highly crucial biological function. Fertility-related proteins also have been proven to be important in many biological entities (i.e., bone marrow and peripheral blood, postnatal mammalian ovary) and parameters (i.e., daily sperm production). As a result, our model can achieve a cross-validation accuracy of 85.8% and an independent accuracy of 91.1%. We also solve the problem of overfitting in the data set by adding dropout layers in the deep learning model. The independent testing results showed sensitivity, specificity, and Matthews correlation coefficient (MCC) values of 90.5%, 91.7%, and 0.82, respectively. Fertility-GRU demonstrates superiority in performance against the state-of-the-art predictor on the same data set. In our proposed study, we provided a method that enables more proteins to be discovered, especially proteins associated with fertility. Moreover, our achievement could promote the use of recurrent networks and gated recurrent units in proteome research. The source code and data set are freely accessible via https://github.com/khanhlee/fertility-gru.
Flavin mono-nucleotides (FMNs) are cofactors that hold responsibility for carrying and transferring electrons in the electron transport chain stage of cellular respiration. Without being facilitated ...by FMNs, energy production is stagnant due to the interruption in most of the cellular processes. Investigation on FMN's functions, therefore, can gain holistic understanding about human diseases and molecular information on drug targets. We proposed a deep learning model using a two-dimensional convolutional neural network and position specific scoring matrices that could identify FMN interacting residues with the sensitivity of 83.7 percent, specificity of 99.2 percent, accuracy of 98.2 percent, and Matthews correlation coefficients of 0.85 for an independent dataset containing 141 FMN binding sites and 1,920 non-FMN binding sites. The proposed method outperformed other previous studies using similar evaluation metrics. Our positive outcome can also promote the utilization of deep learning in dealing with various problems in bioinformatics and computational biology.
•COVID-19 is the period of high uncertainty level.•The stock market's return significantly impacted Bitcoin's return in high uncertainty periods.•The volatility spillover effect from the stock market ...to Bitcoin was found during high uncertainty periods.•No effect of stock market on Bitcoin during low and medium uncertainty periods (or without COVID-19).
This research examined the impact of the stock market on Bitcoin during COVID-19 and other uncertainty periods. Based on the quantile regression results, during periods of high uncertainty, such as COVID-19, the S&P 500 returns significantly affected Bitcoin returns. Moreover, this research applied the VAR (1)–GARCH (1, 1) model to investigate the spillover effect from the stock market to Bitcoin. According to the findings, the shocks from the stock market also influenced Bitcoin's volatility during COVID-19 and other periods of turmoil.
DNA replication is a fundamental task that plays a crucial role in the propagation of all living things on earth. Hence, the accurate identification of its origin could be the key to giving an ...insightful understanding of the regulatory mechanism of gene expression. Indeed, with the robust development of computational techniques and the abundant biological sequencing data, it has become possible for scientists to identify the origin of replication accurately and promptly. This growing concern has drawn a lot of attention among experts in this field. However, to gain better outcomes, more work is required. Therefore, this study is designed to explore the combination of state-of-the-art features and extreme gradient boosting learning system in classifying DNA sequences. Our hybrid approach is able to identify the origin of DNA replication with achieved sensitivity of 85.19%, specificity of 93.83%, accuracy of 89.51%, and MCC of 0.7931. Evidence is presented to show that our proposed method is superior to the state-of-the-art methods on the same benchmark dataset. Moreover, the research results represent a further step towards developing the prediction models for DNA replication in particular and DNA sequences in general.
•A novel method for identifying origin of DNA replication in Saccharomyces cerevisiae with high performance.•Features are extracted from pseudo k-tuple nucleotide composition and continuous bags of nucleotides.•Optimizing hyper-parameters for extreme gradient boosting learning algorithm•Compared with the state-of-the-art methods, our method had a significant improvement in all of the measurement metrics.•A basis for further research that can improve the predictive performance of DNA sequencing problems.
A previous study of the Ca Mau province in Vietnam (Karlsrud et al., 2017a) suggested that ongoing groundwater pumping, which by 2012 had caused a drawdown of the water level in aquifers of up to 20 ...m, caused subsidence of the order 2–4 cm yr−1, and could have reached over 40 cm already. Earlier InSAR studies also suggested ongoing subsidence rates of that order. If the groundwater pumping continues, the total subsidence could reach well over 1 m within the next few decades. The predicted climate driven sea level rise, to be of the order of 60 cm by 2100, will further add to the severe effect of the subsidence. As most of the Ca Mau province lies only 0.5 to 1.5 m a.s.l. (above sea level), the consequences would rapidly become very serious for the livelihood of people in the region. Increased saltwater intrusion into canals and tributaries in the province, and beginning salination of some of the aquifers from which groundwater is pumped, is already observed. In 2017, for the first time, a physical system for subsidence monitoring was installed at three selected locations in the Ca Mau province. At each
location a deep benchmark to a depth of 100 m was installed, each with 3 piezometers at depths ranging from 15 to 60 m. An InSAR corner reflector was also installed at each site. The paper presents data collected from these new monitoring stations up
until the middle 2019. When including estimated subsidence stemming from the soil levels deeper than 100 m, the total present rate of subsidence at the three new monitoring stations range from 17 to 44 mm yr−1. New and previous data show an almost linear decrease in water level within the aquifers from which groundwater is pumped. The data show some seasonal variations in subsidence rates, which is also reflected in variations in pore pressures in the sediments. Such variations are probably related to seasonal variations in levels of groundwater pumping. It is feared that many of the other provinces south of Ho Chi Minh city, face similar subsidence problems. The monitoring program should be extended to verify that. Measures to reduce groundwater and subsidence are urgently needed.
Recently, language representation models have drawn a lot of attention in the natural language processing field due to their remarkable results. Among them, bidirectional encoder representations from ...transformers (BERT) has proven to be a simple, yet powerful language model that achieved novel state-of-the-art performance. BERT adopted the concept of contextualized word embedding to capture the semantics and context of the words in which they appeared. In this study, we present a novel technique by incorporating BERT-based multilingual model in bioinformatics to represent the information of DNA sequences. We treated DNA sequences as natural sentences and then used BERT models to transform them into fixed-length numerical matrices. As a case study, we applied our method to DNA enhancer prediction, which is a well-known and challenging problem in this field. We then observed that our BERT-based features improved more than 5-10% in terms of sensitivity, specificity, accuracy and Matthews correlation coefficient compared to the current state-of-the-art features in bioinformatics. Moreover, advanced experiments show that deep learning (as represented by 2D convolutional neural networks; CNN) holds potential in learning BERT features better than other traditional machine learning techniques. In conclusion, we suggest that BERT and 2D CNNs could open a new avenue in biological modeling using sequence information.
Small bubbles of nitrogen, methane, or argon with an average radius of 50
nm, as measured by scanning electron microscopy, were prepared under atmospheric conditions. The lifetime of the nanobubbles ...extended to more than two weeks. The total amount of gases in the nanobubble solutions reached 600
cm
3 per 1
dm
3 of water, and the liquid density was about 0.988
g/cm
3. The internal pressure of the nanobubbles was estimated to be 6
MPa. The number of nanobubbles was 1.9×10
16 bubbles per 1
dm
3 of water. These findings show that almost no gas samples are dissolved homogeneously in the aqueous solution and that the vast majority is present in the form of nanobubbles, that is, nanobubbles should be thermodynamically unstable. Attenuated total reflectance infrared spectroscopy showed that the surfaces of the nanobubbles contain hard hydrogen bonds that may reduce the diffusivity of gases through the interfacial film.