One of the main problems with the joint use of multiple drugs is that it may cause adverse drug interactions and side effects that damage the body. Therefore, it is important to predict potential drug interactions. However, most of the available prediction methods can only predict whether two drugs interact or not, whereas few methods can predict interaction events between two drugs. Accurately predicting interaction events of two drugs is more useful for researchers to study the mechanism of the interaction of two drugs. In the present study, we propose a novel method, MDF-SA-DDI, which predicts drug–drug interaction (DDI) events based on multi-source drug fusion, multi-source feature fusion and transformer self-attention mechanism. MDF-SA-DDI is mainly composed of two parts: multi-source drug fusion and multi-source feature fusion. First, we combine two drugs in four different ways and input the combined drug feature representation into four different drug fusion networks (Siamese network, convolutional neural network and two auto-encoders) to obtain the latent feature vectors of the drug pairs, in which the two auto-encoders have the same structure, and their main difference is the number of neurons in the input layer of the two auto-encoders. Then, we use transformer blocks that include self-attention mechanism to perform latent feature fusion. We conducted experiments on three different tasks with two datasets. On the small dataset, the area under the precision–recall-curve (AUPR) and F1 scores of our method on task 1 reached 0.9737 and 0.8878, respectively, which were better than the state-of-the-art method. On the large dataset, the AUPR and F1 scores of our method on task 1 reached 0.9773 and 0.9117, respectively. In task 2 and task 3 of two datasets, our method also achieved the same or better performance as the state-of-the-art method. More importantly, the case studies on five DDI events are conducted and achieved satisfactory performance. The source codes and data are available at https://github.com/ShenggengLin/MDF-SA-DDI.

Characterization of the bacterial composition and functional repertoires of microbiome samples is the most common application of metagenomics. Although deep whole-metagenome shotgun sequencing (WMS) provides high taxonomic resolution, it is generally cost-prohibitive for large longitudinal investigations. Until now, 16S rRNA gene amplicon sequencing (16S) has been the most widely used approach and usually cooperates with WMS to achieve cost-efficiency. However, the accuracy of 16S results and its consistency with WMS data have not been fully elaborated, especially by complicated microbiomes with defined compositional information. Here, we constructed two complex artificial microbiomes, which comprised more than 60 human gut bacterial species with even or varied abundance. Utilizing real fecal samples and mock communities, we provided solid evidence demonstrating that 16S results were of poor consistency with WMS data, and its accuracy was not satisfactory. In contrast, shallow whole-metagenome shotgun sequencing (shallow WMS, S-WMS) with a sequencing depth of 1 Gb provided outputs that highly resembled WMS data at both genus and species levels and presented much higher accuracy taxonomic assignments and functional predictions than 16S, thereby representing a better and cost-efficient alternative to 16S for large-scale microbiome studies.

The gut microbes play important roles in human longevity and the gut microbiota profile of centenarians shows some unique features from young adults. Nowadays, most microbial studies on longevity are commonly based on metagenomic sequencing which may lose information about the functional microbes with extremely low abundance. Here, we combined in-depth metagenomic sequencing and large-scale culturomics to reveal the unique gut microbial structure of a Chinese longevity population, and to explore the possible relationship between intestinal microbes and longevity. Twenty-five healthy Hainan natives were enrolled in the study, including 12 centenarians and 13 senior neighbors. An average of 51.1 Gb raw sequencing data were obtained from individual fecal sample. We assembled 1778 non-redundant metagenomic assembled genomes (MAGs), 33.46% of which cannot be classified into known species. Comparison with the ordinary people in Hainan province, the longevous cohort displayed significantly decreased abundance of butyrate-producing bacteria and largely increased proportion of Escherichia coli, Desulfovibrio piger and Methanobrevibacter smithii. These species showed a constant change with aging. We also isolated 8,030 strains from these samples by large-scale culturomics, most of which belonged to 203 known species as identified by MALDI-TOF. Surprisingly, only 42.17% of the isolated species were also detected by metagenomics, indicating obvious complementarity between these two approaches. Combination of two complement methods, in-depth metagenomic sequencing and culturomics, provides deeper insights into the longevity-related gut microbiota. The uniquely enriched gut microbes in Hainan extreme decades population may help to promote health and longevity.

Neuropeptides acting as signaling molecules in the nervous system of various animals play crucial roles in a wide range of physiological functions and hormone regulation behaviors. Neuropeptides offer many opportunities for the discovery of new drugs and targets for the treatment of neurological diseases. In recent years, there have been several data-driven computational predictors of various types of bioactive peptides, but the relevant work about neuropeptides is little at present. In this work, we developed an interpretable stacking model, named NeuroPpred-Fuse, for the prediction of neuropeptides through fusing a variety of sequence-derived features and feature selection methods. Specifically, we used six types of sequence-derived features to encode the peptide sequences and then combined them. In the first layer, we ensembled three base classifiers and four feature selection algorithms, which select non-redundant important features complementarily. In the second layer, the output of the first layer was merged and fed into logistic regression (LR) classifier to train the model. Moreover, we analyzed the selected features and explained the feasibility of the selected features. Experimental results show that our model achieved 90.6% accuracy and 95.8% AUC on the independent test set, outperforming the state-of-the-art models. In addition, we exhibited the distribution of selected features by these tree models and compared the results on the training set to that on the test set. These results fully showed that our model has a certain generalization ability. Therefore, we expect that our model would provide important advances in the discovery of neuropeptides as new drugs for the treatment of neurological diseases.

In the burgeoning microbiome field, powerful sequencing approaches and accompanied bioanalytical methods have made tremendous contributions to the discoveries of breakthroughs, which favor to unravel the intimate interplay between gut microbiota and human health. The proper preservation of samples before being processed is essential to guarantee the authenticity and reliability of microbiome studies. Hence, the development of preservation methods is extremely important to hold samples eligible for the consequent analysis, especially population cohort-based investigations or those spanning species or geography, which frequently facing difficulties in suppling freezing conditions. Although there are several commercial products available, the exploration of cost-efficient and ready-to-use preservation methods are still in a large demand. Here, we performed shotgun metagenomic sequencing and demonstrated that microbial consortia in human fecal samples were substantially preserved within a temporary storage of 4 h, independent of the storage temperature. We also verified a previous reported self-made preservation buffer (PB buffer) could not only preserve fecal microbiota at room temperature up to 4 weeks but also enable samples to endure a high temperature condition which mimics temperature variations in summer logistics. Moreover, PB buffer exhibited suitability for human saliva as well. Collectively, PB buffer may be a valuable choice to stabilize samples if neither freezing facilities nor liquid nitrogen is available.