No abstract

To improve the diagnosis and classification of Alzheimer’s disease (AD), a modeling method is proposed based on the combining magnetic resonance images (MRI) brain structural data with metabolite levels of the frontal and parietal regions. First, multi-atlas brain segmentation technology based on T1-weighted images and edited magnetic resonance spectroscopy (MRS) were used to extract data of 279 brain regions and levels of 12 metabolites from regions of interest (ROIs) in the frontal and parietal regions. The t-test combined with false discovery rate (FDR) correction was used to reduce the dimensionality in the data, and MRI structural data of 54 brain regions and levels of 4 metabolites that obviously correlated with AD were screened out. Lastly, the stacked auto-encoder neural network (SAE) was used to classify AD and healthy controls (HCs), which judged the effect of classification method by fivefold cross validation. The results indicated that the mean accuracy of the five experimental model increased from 96 to 100%, the AUC value increased from 0.97 to 1, specificity increased from 90 to 100%, and F1 value increased from 0.97 to 1. Comparing the effect of each metabolite on model performance revealed that the gamma-aminobutyric acid (GABA) + levels in the parietal region resulted in the most significant improvement in model performance, with the accuracy rate increasing from 96 to 98%, the AUC value increased from 0.97 to 0.99 and the specificity increasing from 90 to 95%. Moreover, the GABA + levels in the parietal region was significantly correlated with Mini Mental State Examination (MMSE) scores of patients with AD (r = 0.627), and the F statistics were largest (F = 25.538), which supports the hypothesis that dysfunctional GABAergic system play an important role in the pathogenesis of AD. Overall, our findings support that a comprehensive method that combines MRI structural and metabolic data of brain regions can improve model classification efficiency of AD.

The domesticated silkworm, Bombyx mori, is an important model system for the order Lepidoptera. Currently, based on third-generation sequencing, the chromosome-level genome of Bombyx mori has been released. However, its transcripts were mainly assembled by using short reads of second-generation sequencing and expressed sequence tags which cannot explain the transcript profile accurately. Here, we used PacBio Iso-Seq technology to investigate the transcripts from 45 developmental stages of Bombyx mori. We obtained 25,970 non-redundant high-quality consensus isoforms capturing ∼60% of previous reported RNAs, 15,431 (∼47%) novel transcripts, and identified 7,253 long non-coding RNA (lncRNA) with a large proportion of novel lncRNA (∼56%). In addition, we found that transposable elements (TEs) exonization account for 11,671 (∼45%) transcripts including 5,980 protein-coding transcripts (∼32%) and 5,691 lncRNAs (∼79%). Overall, our results expand the silkworm transcripts and have general implications to understand the interaction between TEs and their host genes. These transcripts resource will promote functional studies of genes and lncRNAs as well as TEs in the silkworm.

Long noncoding RNAs (lncRNAs), which have a length longer than 200bp (base pair), participate in various critical biological processes. Moreover, they have many similar features with another kind of RNA - coding RNA, such as long length of transcript and poly-A tail. Therefore, distinguish lncRNA and coding RNA can be one important task in bioinformatics. With the advanced and outstanding ability of machine learning, the computational method provides new insight into lncRNA classification. In this study, two feature selection methods (lasso and PCA) are applied to reduce dimension. 8 differentiated features are extracted, and lasso selection indicates better performance than the PCA method. To achieve an advanced performance of lncRNA classification, one novel ensemble learning based on primary learner and secondary learner is constructed. After comparing different kinds of models, ensemble learning achieves the most outstanding performance in AUC and accuracy within the test dataset (The median of Accuracy=0.950228, AUC=0.979664), which may shed light on the classification of lncRNA.