Search Results (35)

Codon optimization is widely used to improve heterologous gene expression in Escherichia coli. However, many existing methods focus primarily on maximizing the codon adaptation index (CAI) and neglect broader aspects of biological context. In this study, we present ColiFormer, a transformer-based codon optimization framework fine-tuned on 3676 high-expression E. coli genes curated from the NCBI database. Built on the CodonTransformer BigBird architecture, ColiFormer employs self-attention mechanisms and a mathematical optimization method (the augmented Lagrangian approach) to balance multiple biological objectives simultaneously, including CAI, GC content, tRNA adaptation index (tAI), RNA stability, and minimization of negative cis-regulatory elements. Based on in silico evaluations on 37,053 native E. coli genes and 80 recombinant protein targets commonly used in industrial studies, ColiFormer demonstrated significant improvements in CAI and tAI values, maintained GC content within biologically optimal ranges, and reduced inhibitory cis-regulatory motifs compared with established codon optimization approaches, while maintaining competitive runtime performance. These results represent computational predictions derived from standard in silico metrics; future experimental work is anticipated to validate these computational predictions in vivo. ColiFormer has been released as an open-source tool alongside the benchmark datasets used in this study. Full article

Yakutia, the largest breeding ground for wild migratory birds in Northeastern Siberia, plays a big role in the global ecology of avian influenza viruses (AIVs). In this study, we present the results of virological surveillance conducted between 2018 and 2023, analyzing 1970 cloacal swab samples collected from 56 bird species. We identified 74 AIVs of H3N6, H3N8, H4N6, H5N3, H7N7, H10N3, and H11N9 subtypes in Anseriformes order. Phylogenetic analysis showed that the isolates belong to the Eurasian lineage and have genetic similarities with strains from East Asia, Europe, and North America. Cluster analysis has demonstrated the circulation of stable AIV genotypes for several years. We assume that Yakutia is an important territory for viral exchange on the migratory routes of migrating birds. In addition, several amino acid substitutions have been found to be associated with increased virulence and adaptation to mammalian hosts, highlighting the potential risk of interspecific transmission. These results provide a critical insight into the ecology of the AIV and highlight the importance of continued monitoring in this geographically significant region. Full article

The study discloses the application of transformer-based deep learning models for the task of super-enhancers prediction in human tumor cell lines with a specific focus on sequence-only features within studied entities of super-enhancer and enhancer elements in the human genome. The proposed SE-prediction method included the GENA-LM application at handling long DNA sequences with the classification task, distinguishing super-enhancers from enhancers using H3K36me, H3K4me1, H3K4me3 and H3K27ac landscape datasets from HeLa, HEK293, H2171, Jurkat, K562, MM1S and U87 cell lines. The model was fine-tuned on relevant sequence data, allowing for the analysis of extended genomic sequences without the need for epigenetic markers as proposed in early approaches. The study achieved balanced accuracy metrics, surpassing previous models like SENet, particularly in HEK293 and K562 cell lines. Also, it was shown that super-enhancers frequently co-localize with epigenetic marks such as H3K4me3 and H3K27ac. Therefore, the attention mechanism of the model provided insights into the sequence features contributing to SE classification, indicating a correlation between sequence-only features and mentioned epigenetic landscapes. These findings support the potential transformer models use in further genomic sequence analysis for bioinformatics applications in enhancer/super-enhancer characterization and gene regulation studies. Full article

With the rapid development of prominent models, NL2SQL has made many breakthroughs, but customers still hope that the accuracy of NL2SQL can be continuously improved through optimization. The method based on large models has brought revolutionary changes to NL2SQL. This paper innovatively proposes a new NL2SQL method based on a large language model (LLM), which could be adapted to an edge-cloud computing platform. First, natural language is converted into Python language, and then SQL is generated through Python. At the same time, considering the traceability characteristics of financial industry regulatory requirements, this paper uses the open-source big model DeepSeek. After testing on the BIRD dataset, compared with most NL2SQL models based on large language models, EX is at least 2.73% higher than the original method, F1 is at least 3.72 higher than the original method, and VES is 6.34% higher than the original method. Through this innovative algorithm, the accuracy of NL2SQL in the financial industry is greatly improved, which can provide business personnel with a robust database access mode. Full article

This study examines how sentiment analysis of environmental, social, and governance (ESG) news affects the financial performance of companies in innovative sectors such as mobility, technology, and renewable energy. Using approximately 9828 general ESG articles from Google News and approximately 140,000 company-specific ESG articles, we performed term frequency-inverse document frequency (TF-IDF) analysis to identify key ESG-related terms and visualize their materiality across industries. We then applied models such as bidirectional encoder representations from transformers (BERT), the robustly optimized BERT pretraining approach (RoBERTa), and big bidirectional encoder representations from transformers (BigBird) for multiclass sentiment analysis, and distilled BERT (DistilBERT), a lite BERT (ALBERT), tiny BERT (TinyBERT), and efficiently learning an encoder that classifies token replacements accurately (ELECTRA) for positive and negative sentiment identification. Sentiment analysis results were correlated with profitability, cash flow, and stability indicators over a three-year period (2019–2021). ESG ratings from Morgan Stanley Capital International (MSCI), a prominent provider that evaluates companies’ sustainability practices, further enriched our analysis. The results suggest that sentiment impacts financial performance differently across industries; for example, positive sentiment correlates with financial success in mobility and renewable energy, while consumer goods often show positive sentiment even with low environmental ESG scores. The study highlights the need for industry-specific ESG strategies, especially in dynamic sectors, and suggests future research directions to improve the accuracy of ESG sentiment analysis. Full article

The aim of this study was to evaluate carcass quality and analyze gastrointestinal functional status, ochratoxin A (OTA) accumulation in tissues and organs, and the health status of turkeys fed diets contaminated with OTA and supplemented with synbiotic preparations in comparison with commercial probiotic feed additives. The research involved 120 female BIG 6 turkeys, divided into six treatment groups (five replicates, four birds per replicate). Wheat naturally contaminated with OTA (662.03 μg/kg) was used in turkey diets. Turkeys in group 1 received an OTA-contaminated diet without additives. Groups 2 and 3 received 0.4 g/kg of probiotic preparation BioPlus 2B or Cylactin. Groups 4, 5, and 6 received 0.5 g/kg of synbiotics S1, S2, or S3. The following parameters were monitored: growth performance, carcass quality, gastrointestinal tract structure and digesta pH, health status, and concentrations of OTA in the blood and tissues of turkeys. The study found no significant differences in the growth performance and carcass quality of turkey. However, the introduction of probiotics or synbiotics into OTA-contaminated feed mixtures resulted in a reduced pH of the digesta in certain sections of the turkey digestive tract (p < 0.05). Additionally, the tested synbiotic additives significantly reduced liver weight in turkeys at weeks 6 and 15 (p < 0.05). The addition of probiotic and synbiotic preparations based on lactic acid bacteria strains, inulin, and S. cerevisiae yeasts to OTA-contaminated diets in commercial turkey farming may improve health status (p < 0.05) and reduce mycotoxin accumulation in organs and tissues of poultry (p < 0.05). Full article

Background: As large language models continue to expand in size and diversity, their substantial potential and the relevance of their applications are increasingly being acknowledged. The rapid advancement of these models also holds profound implications for the long-term design of stimulus-responsive materials used in drug delivery. Methods: The large model used Hugging Face’s Transformers package with BigBird, Gemma, and GPT NeoX architectures. Pre-training used the PubChem dataset, and fine-tuning used QM7b. Chemist instruction training was based on Direct Preference Optimization. Drug Likeness, Synthetic Accessibility, and PageRank Scores were used to filter molecules. All computational chemistry simulations were performed using ORCA and Time-Dependent Density-Functional Theory. Results: To optimize large models for extensive dataset processing and comprehensive learning akin to a chemist’s intuition, the integration of deeper chemical insights is imperative. Our study initially compared the performance of BigBird, Gemma, GPT NeoX, and others, specifically focusing on the design of photoresponsive drug delivery molecules. We gathered excitation energy data through computational chemistry tools and further investigated light-driven isomerization reactions as a critical mechanism in drug delivery. Additionally, we explored the effectiveness of incorporating human feedback into reinforcement learning to imbue large models with chemical intuition, enhancing their understanding of relationships involving -N=N- groups in the photoisomerization transitions of photoresponsive molecules. Conclusions: We implemented an efficient design process based on structural knowledge and data, driven by large language model technology, to obtain a candidate dataset of specific photoswitchable molecules. However, the lack of specialized domain datasets remains a challenge for maximizing model performance. Full article

Balancing food production and biodiversity conservation is a big challenge around the world. Eco-friendly agriculture has the potential to overcome these challenges and achieve sustainability. Although some studies show the high valuation of flagship species (e.g., attractive birds and fish) in paddy land where eco-friendly rice is cultivated, limited research has been performed concerning non-specific species such as generalists inhabiting rice paddies that also contribute to agro-biodiversity and ecosystem services. Therefore, this study assesses the valuation of non-flagship vertebrates and invertebrates by applying a choice experiment to vertebrate- and invertebrate-friendly rice. To understand the spatial/regional heterogeneity of the valuation, a questionnaire survey was distributed to both urban and rural citizens in Japan. Our results demonstrated that almost all respondents expressed a desire to protect both vertebrates and invertebrates, with more appreciation for vertebrates than for invertebrates. The analysis also found regional heterogeneity between urban and rural areas in terms of vertebrate and invertebrate evaluations and purchasing intentions. Our findings indicate marketing potential in Japan to promote eco-friendly rice production in relation to vertebrate and invertebrate conservation. Full article

Despite significant advancements in Artificial Intelligence (AI) and Large Language Models (LLMs), detecting and mitigating bias remains a critical challenge, particularly on social media platforms like X (formerly Twitter), to address the prevalent cyberbullying on these platforms. This research investigates the effectiveness of leading LLMs in generating synthetic biased and cyberbullying data and evaluates the proficiency of transformer AI models in detecting bias and cyberbullying within both authentic and synthetic contexts. The study involves semantic analysis and feature engineering on a dataset of over 48,000 sentences related to cyberbullying collected from Twitter (before it became X). Utilizing state-of-the-art LLMs and AI tools such as ChatGPT-4, Pi AI, Claude 3 Opus, and Gemini-1.5, synthetic biased, cyberbullying, and neutral data were generated to deepen the understanding of bias in human-generated data. AI models including DeBERTa, Longformer, BigBird, HateBERT, MobileBERT, DistilBERT, BERT, RoBERTa, ELECTRA, and XLNet were initially trained to classify Twitter cyberbullying data and subsequently fine-tuned, optimized, and experimentally quantized. This study focuses on intersectional cyberbullying and multilabel classification to detect both bias and cyberbullying. Additionally, it proposes two prototype applications: one that detects cyberbullying using an intersectional approach and the innovative CyberBulliedBiasedBot that combines the generation and detection of biased and cyberbullying content. Full article

Identification of different species of animals has become an important issue in biology and ecology. Ornithology has made alliances with other disciplines in order to establish a set of methods that play an important role in the birds’ protection and the evaluation of the environmental quality of different ecosystems. In this case, the use of machine learning and deep learning techniques has produced big progress in birdsong identification. To make an approach from AI-IoT, we have used different approaches based on image feature comparison (through CNNs trained with Imagenet weights, such as EfficientNet or MobileNet) using the feature spectrogram for the birdsong, but also the use of the deep CNN (DCNN) has shown good performance for birdsong classification for reduction of the model size. A 5G IoT-based system for raw audio gathering has been developed, and different CNNs have been tested for bird identification from audio recordings. This comparison shows that Imagenet-weighted CNN shows a relatively high performance for most species, achieving 75% accuracy. However, this network contains a large number of parameters, leading to a less energy efficient inference. We have designed two DCNNs to reduce the amount of parameters, to keep the accuracy at a certain level, and to allow their integration into a small board computer (SBC) or a microcontroller unit (MCU). Full article

For a long time, it was considered that entomogamy was the only pollination mechanism in the Mediterranean area. However, data recorded in this review prove that ornithogamy and saurogamy also take place. With the exception of the nectarivorous Cinnyris osea (Nectariniidae) which pollinates the mistletoe Picosepalus acaciae in Israel, all birds responsible for the pollination of several plant species in this area are primarily insectivorous, sedentary, or migrating passerine birds, particularly Sylvia atricapilla, S. melanocephala, Phylloscopus collibita and Parus caeruleus. They contribute, together with insects, to the pollination of Anagyris foetida, three species of Scrophularia with big flowers, Rhamnus alaternus, Brassica oleracea, and some other plants. The lacertid lizard Podarcis lilfordi acts as a pollinating agent on several W Mediterranean islands, where it effectively pollinates Euphorbia dendroides, Cneorum tricocum, and presumably Rosmarinus officinalis and Chrithmum maritimum. The flowers of some other plant species are visited by birds or by Podarcis species in the Mediterranean area, where they could also contribute to their pollination. Full article

The Brown-backed Parrotlet, Touit melanonotus, is a rare endemic bird to the Brazilian Atlantic Forest, currently considered as “Vulnerable” in the Brazilian Red List of Threatened Species. We estimated the sex ratio of a wild flock of T. melanonotus using molecular markers, examined morphological variation in 34 museum specimens to test for sexual dimorphism, and conducted a literature review about sex ratio in Psittacidae for comparative purposes. We found a sex ratio of 0.8:1 (male/female; n = 29) in T. melanonotus, and a χ2 Goodness-of-fit test showed no significant difference from equality (p > 0.05). We describe three main categories in plumage: the first (and most common) comprises uniformly lime green birds, slightly darker on the head. The second is composed of individuals who are overall lighter, with the breast feathers washed with light greenish gray, and feathers of the head being dark lime green, presenting a sharp contrast with the breast feathers. The third and the rarest one is composed of birds with light greenish gray underparts with emerald green and darker upper parts. T. melanonotus has no apparent sexual dimorphism. We found no evidence of geographic variation. Sex ratio deviation may not be a parameter increasing the vulnerability of the species. Data like these represent a big leap in the knowledge of the species and has the potential to help and inform conservation efforts. Full article

The study aimed to determine the effect of L-carnitine and Bio-Mos administration on selected production performance, slaughter parameters, elemental and mineral content of liver, breast and thigh muscles, and physical, morphometric, strength and bone mineral composition parameters of turkeys. The experiment was conducted on 360 six-week-old Big-6 turkey females, randomly divided into three groups of 120 birds each (six replicates of 20 birds). The turkeys of the control group were fed standard feed without additives; group II was fed with drinking water, a preparation containing L-carnitine at a dose of 0.83 mL/L, while group III was provided mixed feed with 0.5% Bio-Mos. The addition of L-carnitine and Bio-Mos increased body weight at 16 weeks (p = 0.047) and reduced the proportion of fat in the breast muscle (p = 0.029) and liver (p = 0.027). It also modified the content of some minerals in breast muscle, thigh muscle, liver, and bone. Furthermore, the addition of L-carnitine and Bio-Mos increased bone mass and length and modified the value of selected morphometric and strength parameters. The results indicate a positive effect of the applied feed additives on selected rearing indices and carcass quality while improving the elasticity and fracture toughness of the femur. There is a need for further research to determine optimal doses of L-carnitine and Bio-Mos in poultry nutrition. Full article

Big musical events often coincide with natural spaces, and therefore they may have an impact on sensitive ecosystems. Here, a story of events that took place following a big event on an Italian beach within a Special Protection Area (SPA; hosting embryonic shifting dunes and plover birds of conservation concern) is reported. Following a theatrical approach to conservation, this story unfolds in three acts (Act I: The Premises; II: On the Field; III: Long-Term Effects) that include the social targets (‘actors’) involved (i.e., the pop star’s staff, ONG, institutions, and local stakeholders) as well as the critical issues and conflicts. This experience provides some conservation lessons: (i) big musical events can have an impact on sensitive socio-ecosystems; (ii) the intrinsic value of coastal ecosystems has been underestimated since the site selection was carried out by decision makers with inaccurate/inappropriate use of digital tools; (iii) communication among the private organizers, public institutions, ONG, and people was poor; (iv) the availability of huge economic resources has made the local municipality vulnerable; (v) digital social processes increased polarization between opposing parties with an increase in local conflicts among Public Agencies; and (vi) these conflicts had long-term cascade effects on the nature reserve’s management. To communicate conservation stories, I encourage conservation practitioners to use a theatrical approach to communicate local events with socio-ecological implications—increasing awareness of human–wildlife conflicts and cognitive bias emerging after unshared decisions—using simplified conceptual frameworks. Full article

In order to increase growth performance and carcass composition, including meat quality, as demanded by modern customers, alternative feed additives are sought after as a result of the withdrawal of antibiotics employed as growth promoters in poultry feeding. Therefore, we conducted a study to see how effective three feed additives added to the diets and water of turkey hens were. The experiment consisted of 200 Big 6 turkey hens divided into two equinumerous groups (C and E), with five subgroups in each. The 14-week-long growth performance study comprised five feeding periods. Both groups of birds were fed complete feed rations with mineral and vitamin supplements. The factor differentiating the groups were effective microorganisms and Humokarbowit added to the birds’ diets and garlic extract added to the drinking water in the experimental group (E) only. It was demonstrated that the included feed additives in the diets and water of turkey hens significantly increased (by 10%) the FBW and decreased (by 14%) the FCR throughout the rearing period. Birds from the E group scored significantly higher (by 3.6%) on the dressing percentage, and their muscularity and fattening grade were improved. Turkey fed rations containing the evaluated feed additives had a smaller share of the gastrointestinal tract in the body weight and a shorter duodenum and caecum (p ≤ 0.05). The muscles of turkey hens from group E featured a lower pH₂₄ and were of lighter colour (p ≤ 0.05). To sum up, the use of effective microorganisms and Humokarbowit in the diets and garlic extract in the drinking water of turkey hens should be recommended in view of improved growth performance and carcass composition. Full article