It is well known that DNA sequence contains a certain amount of transcription factors (TF) binding sites, and only part of them are identified through biological experiments. However, these ...experiments are expensive and time-consuming. To overcome these problems, some computational methods, based on k-mer features or convolutional neural networks, have been proposed to identify TF binding sites from DNA sequences. Although these methods have good performance, the context information that relates to TF binding sites is still lacking. Research indicates that standard recurrent neural networks (RNN) and its variants have better performance in time-series data compared with other models. In this study, we propose a model, named KEGRU, to identify TF binding sites by combining Bidirectional Gated Recurrent Unit (GRU) network with k-mer embedding. Firstly, DNA sequences are divided into k-mer sequences with a specified length and stride window. And then, we treat each k-mer as a word and pre-trained word representation model though word2vec algorithm. Thirdly, we construct a deep bidirectional GRU model for feature learning and classification. Experimental results have shown that our method has better performance compared with some state-of-the-art methods. Additional experiments about embedding strategy show that k-mer embedding will be helpful to enhance model performance. The robustness of KEGRU is proved by experiments with different k-mer length, stride window and embedding vector dimension.
This article comments on:
Steven D. Johnson, Jeremy J. Midgley and Nicola Illing. The enantiostylous floral polymorphism of Barberetta aurea (Haemodoraceae) facilitates wing pollination by syrphid ...flies, Annals of Botany, Volume 132, Issue 6, 3 November 2023, Pages 1107–1118, https://doi.org/10.1093/aob/mcad118
• Visual signals attractive to friends may also attract enemies. The bright colors of anthers and pollen have generally been thought to attract pollinators. We hypothesize that visual crypsis of ...anthers, and particularly pollen, should be favored in flowering plants because protection from pollen collectors reduces the loss of male gametes.
• To understand adaptive strategies relating to the color of pollen, we measured the color of pollen, undehisced anther sacs, and their background, the corolla, with a spectrometer for 104 insect-pollinated flowering species from a natural community in Hengduan Mountains, southwest China.
• The colors of anthers, pollen and corollas were diverse in these species. The color diversity of exposed pollen was significantly higher than that of concealed pollen (i.e. where anthers are enclosed or shielded by corollas). The color contrast between pollen and corolla was significantly smaller in species with exposed pollen than in those with concealed pollen. Unlike anther color, exposed pollen color tended to match its background corolla color.
• Our phylogenetic comparative analysis showed contrasting effects of pollen color patterns between flowers with exposed pollen and those with concealed pollen, revealing a strategy of hiding pollen from pollen thieves via visual crypsis.
Abstract
DNA/RNA motif mining is the foundation of gene function research. The DNA/RNA motif mining plays an extremely important role in identifying the DNA- or RNA-protein binding site, which helps ...to understand the mechanism of gene regulation and management. For the past few decades, researchers have been working on designing new efficient and accurate algorithms for mining motif. These algorithms can be roughly divided into two categories: the enumeration approach and the probabilistic method. In recent years, machine learning methods had made great progress, especially the algorithm represented by deep learning had achieved good performance. Existing deep learning methods in motif mining can be roughly divided into three types of models: convolutional neural network (CNN) based models, recurrent neural network (RNN) based models, and hybrid CNN–RNN based models. We introduce the application of deep learning in the field of motif mining in terms of data preprocessing, features of existing deep learning architectures and comparing the differences between the basic deep learning models. Through the analysis and comparison of existing deep learning methods, we found that the more complex models tend to perform better than simple ones when data are sufficient, and the current methods are relatively simple compared with other fields such as computer vision, language processing (NLP), computer games, etc. Therefore, it is necessary to conduct a summary in motif mining by deep learning, which can help researchers understand this field.
Premise
Co‐flowering species that have not evolved an avoidance mechanism may have tolerance to heterospecific pollen (HP) deposition as an adaptive strategy to minimize any deleterious effects of HP ...transfer, but empirical evidence for the tolerance hypothesis remains scarce.
Methods
To estimate the potential effects of heterospecific pollen deposition (HPD) on female reproductive success, we counted conspecific (CP) and HP pollen grains deposited on stigmas and assessed subsequent seed set of both open‐ and hand‐pollinated flowers in three co‐flowering Silene species with exposed stigmas that usually received numerous HP grains on the elongated receptive area.
Results
The percentage of HP grains per flower (HP%) varied from 16.6% to 43.0% among three species. Silene chungtienensis had lower HP%, and the CP‐HP relationship was neutral; S. gracilicaulis and S. yunnanensis had a relatively higher HP% with a positive CP‐HP relationship. The effects of CP and HP number on natural seed set were positive for all three species, but HP% had stronger negative effects in S. chungtienensis and S. gracilicaulis. In hand‐pollinated flowers of the three Silene species, seed set did not decrease with HP whether CP was in excess or insufficient, indicating no negative effects of HPD on seed production.
Conclusions
Consistent with the tolerance hypothesis, our results indicated that species with higher HP interference are likely to be tolerant to an increase in HP%. These species with generalist‐pollinated flowers and exposed large stigmas may benefit from an increase of conspecific pollen deposition, despite the associated increase in heterospecific pollen deposition.
Premise
Why have pollen grains evolved to be exceptionally large in some species? Pollen‐feeding hypothesis suggests that if the proportion of pollen amounts for feeding is reduced in a flower, the ...low allocation to pollen number would allow pollen grains to be larger.
Methods
To examine whether species with large pollen grains experience low pollen consumption, the behavior of insects feeding on nectar and pollen was observed and pollen transfer efficiency was estimated for four visitor types in Geranium delavayi. To see whether bees actively collected pollen, the numbers of grains in pollen baskets and on the body were compared. Both nutritional value (total protein and lipid) and chemical defense (phenolic metabolites) in pollen against pollen feeders were measured.
Results
Bumblebees and honeybees foraged for nectar, rarely groomed pollen into corbiculae, and had >5× higher pollen transfer efficiency than smaller solitary bees and flies, which were pollen eaters that removed more pollen but deposited less. Pollen grains were characterized by low protein and high lipid content with a low protein‐lipid ratio, an unfavorable combination for bumblebees. Three secondary metabolites were significantly higher in pollen grains (7.77 mg/g) than in petals (1.08 mg/g) or in nectar (0.44 mg/g), suggesting stronger chemical defense in pollen.
Conclusions
Our results indicated that large bees took nectar but little of the nutritionally poor and highly toxic pollen. These data support one prediction of the pollen‐feeding hypothesis, that species with few and large pollen grains would also have low pollen‐consumption rates.
In this brief, a novel local descriptor, named local binary count (LBC), is proposed for rotation invariant texture classification. The proposed LBC can extract the local binary grayscale difference ...information, and totally abandon the local binary structural information. Although the LBC codes do not represent visual microstructure, the statistics of LBC features can represent the local texture effectively. In addition, a completed LBC (CLBC) is also proposed to enhance the performance of texture classification. Experimental results obtained from three databases demonstrate that the proposed CLBC can achieve comparable accurate classification rates with completed local binary pattern.
Objectives
Chronic rhinosinusitis (CRS) reduces the health‐related quality of life and subsequently causes a tremendous socio‐economic impact. Although many studies have been conducted, few have ...identified a relationship between bacteriological characteristics and different phenotypes or endotypes. Therefore, this study aimed to elucidate the recent trends in bacterial cultures from different types of CRS in the Asian population.
Methods
This retrospective case–control study recruited patients diagnosed with CRS who underwent functional endoscopic sinus surgery (FESS) at a tertiary hospital in Taiwan. The patients were classified into those with chronic rhinosinusitis with nasal polyps (CRSwNP)/chronic rhinosinusitis without nasal polyps (CRSsNP), eosinophilic chronic rhinosinusitis (eCRS)/non‐eosinophilic chronic rhinosinusitis (NECRS), and central compartment atopic disease (CCAD)/lateral‐dominant nasal polyp (LDNP) groups. The demographic data and bacteriological characteristics of the groups were analyzed.
Results
We included 503 patients, identifying no significant difference between CRSwNP and CRSsNP for several common bacteria in CRS. The number of Staphylococcus epidermidis isolates in culture was significantly higher in the NECRS group (50.46% vs. 32.56%, p = 0.0003) than that in the eCRS group. The number of methicillin‐resistant Staphylococcus aureus (MRSA; 8.51% vs. 2.35%, p = 0.0221) positive isolates was significantly higher in the CCAD group than that in the LDNP group.
Conclusions
This was the first study in Asia to analyze the relationship between bacteriological characteristics and CCAD. MRSA is significantly higher in the CCAD group than that in the LDNP group. Recognizing the unique microbiology of CRSwNP, eCRS, and CCAD is crucial when selecting antimicrobial therapy to lessen the socio‐economic impact.
Level of Evidence
3 Laryngoscope, 134:1071–1076, 2024
Modeling in-vivo protein-DNA binding is not only fundamental for further understanding of the regulatory mechanisms, but also a challenging task in computational biology. Deep-learning based methods ...have succeed in modeling in-vivo protein-DNA binding, but they often (1) follow the fully supervised learning framework and overlook the weakly supervised information of genomic sequences that a bound DNA sequence may has multiple TFBS(s), and, (2) use one-hot encoding to encode DNA sequences and ignore the dependencies among nucleotides. In this paper, we propose a weakly supervised framework, which combines multiple-instance learning with a hybrid deep neural network and uses k-mer encoding to transform DNA sequences, for modeling in-vivo protein-DNA binding. Firstly, this framework segments sequences into multiple overlapping instances using a sliding window, and then encodes all instances into image-like inputs of high-order dependencies using k-mer encoding. Secondly, it separately computes a score for all instances in the same bag using a hybrid deep neural network that integrates convolutional and recurrent neural networks. Finally, it integrates the predicted values of all instances as the final prediction of this bag using the Noisy-and method. The experimental results on in-vivo datasets demonstrate the superior performance of the proposed framework. In addition, we also explore the performance of the proposed framework when using k-mer encoding, and demonstrate the performance of the Noisy-and method by comparing it with other fusion methods, and find that adding recurrent layers can improve the performance of the proposed framework.
Co‐flowering plants may commonly experience interspecific pollination. It remains unknown, however, whether interspecific pollination is a largely stochastic process or consistent enough over years ...to exert selection for traits that can reduce interspecific pollination or ameliorate its deleterious effects on reproduction. To assess the likelihood of this precondition being met, stigmatic pollen loads on 17–34 insect‐pollinated plant species over three consecutive years were scored in a subalpine meadow in southwestern China. Plant species varied significantly in the amount and proportion of heterospecific pollen (HP) on stigmas. Both the number of HP species and the proportion of the pollen load that was HP for each recipient species correlated positively between years (reflected in pairwise correlations for all year‐by‐year combinations). Although inter‐annual variation was smaller for conspecific pollen (CP) than for HP loads, species tended to experience either consistently high or consistently low HP proportions across years. We found that species with higher stigmatic HP proportions generally experienced lower proportional variation in stigmatic HP, an unexpected result if high HP loads are the result of rare stochastic events. The novel finding of between‐year consistency in stigmatic loads of heterospecific pollen suggests that adaption to stigmatic loads of HP is possible, and two divergent strategies may have evolved: HP avoidance and HP tolerance. The observation of temporally consistent differences among species in levels of HP supports the idea that natural selection may be operating either to increase tolerance or to minimize arrival of heterospecific pollen on stigmas in co‐flowering plants. Such adaptations may be important for the maintenance of high levels of local plant diversity in biodiversity hotspots such as our study area.