Search

Since 2015, a fast growing number of deep learning–based methods have been proposed for protein–ligand binding site prediction and many have achieved promising performance. These methods, however, neglect the imbalanced nature of binding site prediction problems. Traditional data-based approaches for handling data imbalance employ linear interpolation of minority class samples. Such approaches may not be fully exploited by deep neural networks on downstream tasks. We present a novel technique for balancing input classes by developing a deep neural network–based variational autoencoder (VAE) that aims to learn important attributes of the minority classes concerning nonlinear combinations. After learning, the trained VAE was used to generate new minority class samples that were later added to the original data to create a balanced dataset. Finally, a convolutional neural network was used for classification, for which we assumed that the nonlinearity could be fully integrated. As a case study, we applied our method to the identification of FAD- and FMN-binding sites of electron transport proteins. Compared with the best classifiers that use traditional machine learning algorithms, our models obtained a great improvement on sensitivity while maintaining similar or higher levels of accuracy and specificity. We also demonstrate that our method is better than other data imbalance handling techniques, such as SMOTE, ADASYN, and class weight adjustment. Additionally, our models also outperform existing predictors in predicting the same binding types. Our method is general and can be applied to other data types for prediction problems with moderate-to-heavy data imbalances. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Improving the relationship between the police and the public is a major focus of American police agencies. Community–police dialogues represent a potential avenue to facilitate positive interactions and promote the four pillars of procedural justice (voice, neutrality, respect, and impartiality). Recent funding has allowed one agency to host six community–police dialogues (three with adult community members and three with youth community members). These events allow citizens to participate in a community building circle with police to build community and discuss various concerns relevant to their city. Using pre- and post-test data from dialogue participants, the current study seeks to determine the effects of the community–police dialogue on community perceptions of voice, trust, respect, and empathy towards their police department. Following participation, community members reported a greater willingness to work with police to improve relationships, greater trust and respect towards police officers, and a greater awareness of the challenges faced by police officers. Some key differences emerged depending on the age group of the participant. Similarly, police officers reported higher levels of respect for and trust in the community and greater feelings of respect from residents. These findings highlight the potential for dialogue events to promote procedural justice and enhance community partnerships, especially for youth community members. [ABSTRACT FROM AUTHOR]

Conversations are an essential form of communication in daily family life. Specific patterns of caregiver–child conversations have been linked to children's socio-cognitive development and child-relationship quality beyond the immediate family environment. Recently, interpersonal neural synchronization has been proposed as a neural mechanism supporting conversation. Here, we present a functional near-infrared spectroscopy (fNIRS) hyperscanning study looking at the temporal dynamics of neural synchrony during mother–child conversation. Preschoolers (20 boys and 20 girls, M age 5;07 years) and their mothers (M age 36.37 years) were tested simultaneously with fNIRS hyperscanning while engaging in a free verbal conversation lasting for 4 min. Neural synchrony (using wavelet transform coherence analysis) was assessed over time. Furthermore, each conversational turn was coded for conversation patterns comprising turn-taking, relevance, contingency and intrusiveness. Results from linear mixed-effects modeling revealed that turn-taking, but not relevance, contingency or intrusiveness predicted neural synchronization during the conversation over time. Results are discussed to point out possible variables affecting parent–child conversation quality and the potential functional role of interpersonal neural synchronization for parent–child conversation. [ABSTRACT FROM AUTHOR]

Female mate choice can result in direct benefits to the female or indirect benefits through her offspring. Females can increase their fitness by mating with males whose genes encode increased survivorship and reproductive output. Alternatively, male investment in enhanced mating success may come at the cost of reduced investment in offspring fitness. Here, we measure male mating success in a mating arena that allows for male–male, male–female and female–female interactions in Drosophila melanogaster. We then use isofemale line population measurements to correlate male mating success with sperm competitive ability, the number of offspring produced and the indirect benefits of the number of offspring produced by daughters and sons. We find that males from populations that gain more copulations do not increase female fitness through increased offspring production, nor do these males fare better in sperm competition. Instead, we find that these populations have a reduced reproductive output of sons, indicating a potential reproductive trade‐off between male mating success and offspring quality. [ABSTRACT FROM AUTHOR]

Polyandrous females allow for sexual selection to persist after mating. In the event that females successfully mate with more than one male, sperm competition can occur. Seminal fluid proteins can indirectly affect a male's success in sperm competition through reducing the remating behaviour of females and can directly influence sperm competition through directly displacing competitor sperm or inducing females to eject it. These direct effects on competitor sperm are thought to contribute to the 'second male advantage', whereby the second male to mate sires the majority of offspring. Here, we show an additional mechanism where seminal proteins already present within a mated female appear to enhance offspring production of later competitor males, and contribute to second male advantage. Counter to expectation, increased offspring production was not due to a priming effect of greater early female productivity, nor was it through a general and consistent increase in offspring production. Instead, enhanced productivity was solely through lengthening the time that offspring are sired by the second male, indicating that seminal proteins from the first male to mate may enhance second male advantage through a presumably unintended protective effect on subsequent competitor sperm. [ABSTRACT FROM AUTHOR]

Motivation Efflux protein plays a key role in pumping xenobiotics out of the cells. The prediction of efflux family proteins involved in transport process of compounds is crucial for understanding family structures, functions and energy dependencies. Many methods have been proposed to classify efflux pump transporters without considerations of any pump specific of efflux protein families. In other words, efflux proteins protect cells from extrusion of foreign chemicals. Moreover, almost all efflux protein families have the same structure based on the analysis of significant motifs. The motif sequences consisting of the same amount of residues will have high degrees of residue similarity and thus will affect the classification process. Consequently, it is challenging but vital to recognize the structures and determine energy dependencies of efflux protein families. In order to efficiently identify efflux protein families with considering about pump specific, we developed a 2 D convolutional neural network (2 D CNN) model called DeepEfflux. DeepEfflux tried to capture the motifs of sequences around hidden target residues to use as hidden features of families. In addition, the 2 D CNN model uses a position-specific scoring matrix (PSSM) as an input. Three different datasets, each for one family of efflux protein, was fed into DeepEfflux, and then a 5-fold cross validation approach was used to evaluate the training performance. Results The model evaluation results show that DeepEfflux outperforms traditional machine learning algorithms. Furthermore, the accuracy of 96.02%, 94.89% and 90.34% for classes A, B and C, respectively, in the independent test results show that our model can perform well and can be used as a reliable tool for identifying families of efflux proteins in transporters. Availability and implementation The online version of deepefflux is available at http://deepefflux.irit.fr. The source code of deepefflux is available both on the deepefflux website and at http://140.138.155.216/deepefflux/. Supplementary information Supplementary data are available at Bioinformatics online. [ABSTRACT FROM AUTHOR]

Accumulation of organic wastes, especially in livestock facilities, can be a potential pollution issue. The black soldier fly, Hermetia illucens L. (Diptera: Stratiomyidae), can consume a wide range of organic material and has the potential to be used in waste management. In addition, the prepupae stage of this insect can be harvested and used as a valuable nutritious feed for animal livestock. Five waste types with a wide range of organic source matter were specifically chosen to evaluate the consumption and reduction ability of black soldier fly larvae. H. illucens was able to reduce all waste types examined: 1) control poultry feed, 2) pig liver, 3) pig manure, 4) kitchen waste, 5) fruits and vegetables, and 6) rendered fish. Kitchen waste had the greatest mean rate of reduction (consumption by black soldier fly) per day and produced the longest and heaviest black soldier flies. Larvae reared on liver, manure, fruits and vegetables, and fish were approximately the same length and weight as larvae fed the control feed, although some diets produced larvae with a higher nutritional content. The black soldier fly has the ability to consume and reduce organic waste and be utilized as valuable animal feed. Exploration of the potential use of black soldier flies as an agent for waste management on a large-scale system should continue. [ABSTRACT FROM AUTHOR]