AbDPP: Target-oriented antibody design with pretraining and prior biological structure knowledge.

Antibodies represent a crucial class of complex protein therapeutics and are essential in the treatment of a wide range of human diseases. Traditional antibody discovery methods, such as hybridoma and phage display technologies, suffer from limitations including inefficiency and a restricted exploration of the immense space of potential antibodies. To overcome these limitations, we propose a novel method for generating antibody sequences using deep learning algorithms called AbDPP (target-oriented antibody design with pretraining and prior biological knowledge). AbDPP integrates a pretrained model for antibodies with biological region information, enabling the effective use of vast antibody sequence data and intricate biological system understanding to generate sequences. To target specific antigens, AbDPP incorporates an antibody property evaluation model, which is further optimized based on evaluation results to generate more focused sequences. The efficacy of AbDPP was assessed through multiple experiments, evaluating its ability to generate amino acids, improve neutralization and binding, maintain sequence consistency, and improve sequence diversity. Results demonstrated that AbDPP outperformed other methods in terms of the performance and quality of generated sequences, showcasing its potential to enhance antibody design and screening efficiency. In summary, this study contributes to the field by offering an innovative deep learning-based method for antibody generation, addressing some limitations of traditional approaches, and underscoring the importance of integrating a specific antibody pretrained model and the biological properties of antibodies in generating novel sequences. The code and documentation underlying this article are freely available at https://github.com/zlfyj/AbDPP.

A novel deep-learning model based on τ-shaped convolutional network (τNet) with long short-term memory (LSTM) for physiological fatigue detection from EEG and EOG signals.

In recent years, fatigue driving has become the main cause of traffic accidents, leading to increased attention towards fatigue detection systems. However, the pooling and strided convolutional operations in fatigue detection algorithm based on traditional deep learning methods may led to the loss of some useful information. This paper proposed a novel τ -shaped convolutional network ( τ Net ) aiming to address this issue. Unlike traditional network structures, τ Net incorporates the operations of upsampling features and concatenating high- and low-level features, enabling full utilization of useful information. Moreover, considering that the fatigue state is a mental state involving temporal evolution, we proposed the novel long short-term memory (LSTM)- τ -shaped convolutional network (LSTM- τ Net ), a parallel structure composed of LSTM and τ Net for fatigue detection, where τ Net extracts time-invariant features with location information, and LSTM extracts long temporal dependencies. We compared LSTM- τ Net with six competing methods based on two datasets. Results showed that the proposed algorithm achieved higher classification accuracy than the other methods, with 94.25% on EEG data (binary classification) and 82.19% on EOG data (triple classification). Additionally, the proposed algorithm exhibits low computational cost, good training stability, and robustness against insufficient training. Therefore, it is promising for further implementation of fatigue online detection systems.

A generative adaptive convolutional neural network with attention mechanism for driver fatigue detection with class-imbalanced and insufficient data.

Over the past few years, fatigue driving has emerged as one of the main causes of traffic accidents, necessitating the development of driver fatigue detection systems. However, many existing methods involves tedious manual parameter tunings, a process that is both time-consuming and results in task-specific models. On the other hand, most of the researches on fatigue recognition are based on class-balanced and sufficient data, and effectively "mine" meaningful information from class-imbalanced and insufficient data for fatigue recognition is still a challenge. In this paper, we proposed two novel models, the attention-based residual adaptive multiscale fully convolutional network-long short term memory network (ARMFCN-LSTM), and the Generative ARMFCN-LSTM (GARMFCN-LSTM) aiming to address this issue. ARMFCN-LSTM excels at automatically extracting multiscale representations through adaptive multiscale temporal convolutions, while capturing temporal dependency features through LSTM. GARMFCN-LSTM integrates Wasserstein GAN with gradient penalty (WGAN-GP) into ARMFCN-LSTM to improve driver fatigue detection performance by alleviating data scarcity and addressing class imbalances. Experimental results show that ARMFCN-LSTM achieves the highest classification accuracy of 95.84% in driver fatigue detection on the class-balanced EEG dataset (binary classification), and GARMFCN-LSTM attained an improved classification accuracy of 84.70% on the class-imbalanced EOG dataset (triple classification), surpassing the competing methods. Therefore, the proposed models are promising for further implementations in online driver fatigue detection systems.

A novel command generation method for SSVEP-based BCI by introducing SSVEP blocking response.

Increasing the number of commands in a steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI) by increasing the number of visual stimuli has been widely studied. This paper proposes a novel BCI paradigm based on SSVEP and SSVEP blocking responses (defined as the disappearance or attenuation of the ongoing SSVEP) to increase the number of BCI commands with limited visual stimuli, in which the duration of SSVEP blocking response can be voluntarily controlled by users. Besides, this paper also proposes a frequency-specific threshold method and a unified threshold method to identify SSVEP blocking response. The paradigm includes a frequency recognition phase and an SSVEP blocking response identification phase. Filter bank canonical correlation analysis is used to detect the stimulation frequency, and the proposed threshold method is used to identify the SSVEP blocking response and calculate the blocking duration. The experimental results show that the two proposed threshold methods can effectively identify the SSVEP blocking response with different blocking duration and alternative stimulation frequencies. When there are Nf stimulation frequencies, the number of commands can be increased to Nf×Nt using the proposed paradigm, where Nt blocking durations correspond to each stimulus. This study demonstrates that the proposed paradigm based on SSVEP and SSVEP blocking responses is effective in increasing the number of BCI commands and has great potential for practical applications.

Toxin-screening and identification of bacteria isolated from highly toxic marine gastropod Nassarius semiplicatus.

Bacteria isolated from a highly toxic sample of gastropod Nassarius semiplicatus in Lianyungang, Jiangsu Province in July 2007, were studied to probe into the relationship between bacteria and toxicity of nassariid gastropod. The toxicity of the gastropod sample was 2 x 10(2)mouse unit (MU) per gram of tissue (wet weight). High concentration of tetrodotoxin (TTX) and its analogues (TTXs) were found in the digestive gland and muscle of the gastropod, using high performance liquid chromatography coupled with mass chromatography (LC-MS). Bacterial strains isolated from the digestive gland were cultured and screened for TTX with a competitive ELISA method. Tetrodotoxin was detected in a proportion of bacterial strains, but the toxin content was low. Partial 16S ribosomal DNA (rDNA) of the TTX-producing strains was then sequenced and compared with those published in the GenBank to tentatively identify the toxic strains. It was found that most of the toxic strains were closely affiliated with genus Vibrio, and the others were related to genus Shewanella, Marinomonas, Tenacibaculum and Aeromonas. These findings suggest that tetrodotoxin-producing bacteria might play an important role in tetrodotoxin accumulation/production in N. semiplicatus.

Bilateral pelvic lymph node dissection for Chinese patients with penile cancer: a multicenter collaboration study.

BACKGROUND: Current guidelines recommend pelvic lymphadenectomy (PLND) for patients with pelvic lymph node metastasis and special state. However, these data and recommendations do not distinguish the role of PLND in different patient groups and confirm the final benefits. The aim of this study was to confirm the efficacy of pelvic lymphadenectomy (PLND) for the different groups of patients. METHODS: Data obtained from 7 centers were retrospectively analyzed. Of the patients, 190 pN2-3 penile carcinoma patients confirmed by bilateral inguinal lymph node excision were included in this study. Sixty-nine and 121 of these patients did and did not undergo bilateral PLND, respectively. The baseline differences from the patients were matched by propensity score analysis. RESULTS: In this study, the Kaplan-Meier estimated disease-specific survival (DSS) was not significantly different between the PLND and no-PLND groups (P = 0.796). According to the propensity score matching for T stage, N stage, grade, adjuvant therapies, and lymph node stage (number of inguinal lymph node metastasis and extranodal extension), 48 patients were selected for each group. Among the pN2 patients, the PLND group showed higher DSS rates than the no-surgery group (P = 0.030). However, even after matching, survival did not differ between the PLND and no-PLND patients among all patients (P = 0.609) and pN3 patients (P = 0.417) with comparable DSS. CONCLUSION: Bilateral PLND may improve survival in pN2 patients. Men with pN3 may not benefit from bilateral PLND.

Development of a new outcome prediction model for Chinese patients with penile squamous cell carcinoma based on preoperative serum C-reactive protein, body mass index, and standard pathological risk factors: the TNCB score group system.

PURPOSE: To determine the predictive value and feasibility of the new outcome prediction model for Chinese patients with penile squamous cell carcinoma. RESULTS: The 3-year disease-specific survival (DSS) survival (DSS) was 92.3% in patients with < 8.70 mg/L CRP and 54.9% in those with elevated CRP (P < 0.001). The 3-year DSS was 86.5% in patients with a BMI < 22.6 Kg/m2 and 69.9% in those with a higher BMI (P = 0.025). In a multivariate analysis, pathological T stage (P < 0.001), pathological N stage (P = 0.002), BMI (P = 0.002), and CRP (P = 0.004) were independent predictors of DSS. A new scoring model was developed, consisting of BMI, CRP, and tumor T and N classification. In our study, we found that the addition of the above-mentioned parameters significantly increased the predictive accuracy of the system of the American Joint Committee on Cancer (AJCC) anatomic stage group. The accuracy of the new prediction category was verified. METHODS: A total of 172 Chinese patients with penile squamous cell cancer were analyzed retrospectively between November 2005 and November 2014. Statistical data analysis was conducted using the nonparametric method. Survival analysis was performed with the log-rank test and the Cox proportional hazard model. Based on regression estimates of significant parameters in multivariate analysis, a new BMI-, CRP- and pathologic factors-based scoring model was developed to predict disease--specific outcomes. The predictive accuracy of the model was evaluated using the internal and external validation. CONCLUSIONS: The present study demonstrated that the TNCB score group system maybe a precise and easy to use tool for predicting outcomes in Chinese penile squamous cell carcinoma patients.