Biology subjects

Pan, X.

Publications and source records attributed to Pan, X..

6 recordsLinked to original sources

bioRxiv2018

D-GPM: a deep learning method for gene promoter methylation inference

BackgroundGene promoter methylation plays a critical role in a wide range of biological processes, such as transcriptional expression, gene imprinting, X chromosome inactivation, etc. Whole-genome bisulfite sequencing generates a comprehensive profiling of the gene methylation levels but is limited by a high cost. Recent studies have partitioned the genes into landmark genes and target genes and suggested that the landmark gene expression levels capture adequate information to reconstruct the target gene expression levels. Moreover, the methylation level of the promoter is usually negatively correlated with its corresponding gene expression. This result inspired us to propose that the methylation level of the promoters might be adequate to reconstruct the promoter methylation level of target genes, which would eventually reduce the cost of promoter methylation profiling.\n\nResultsHere, we developed a deep learning model (D-GPM) to predict the whole-genome promoter methylation level based on the methylation profile of the landmark genes. We benchmarked D-GPM against three machine learning methods, namely, linear regression (LR), regression tree (RT) and support vector machine (SVM), based on two criteria: the mean absolute deviation (MAE) and the Pearson correlation coefficient (PCC). After profiling the methylation beta value (MBV) dataset from the TCGA, with respect to MAE and PCC, we found that D-GPM outperforms LR by 9.59% and 4.34%, RT by 27.58% and 22.96% and SVM by 6.14% and 3.07% on average, respectively. For the number of better-predicted genes, D-GPM outperforms LR in 92.65% and 91.00%, RT in 95.66% and 98.25% and SVM in 85.49% and 81.56% of the target genes.\n\nConclusionsD-GPM acquires the least overall MAE and the highest overall PCC on MBV-te compared to LR, RT, and SVM. For a genewise comparative analysis, D-GPM outperforms LR, RT, and SVM in an overwhelming majority of the target genes, with respect to the MAE and PCC. Most importantly, D-GPM predominates among the other models in predicting a majority of the target genes according to the model distribution of the least MAE and the highest PCC for the target genes.

bioinformatics↗

bioRxiv2018

Osteoblastic PLEKHO1 contributes to joint inflammation in rheumatoid arthritis

Osteoblasts participating in the inflammation regulation gradually obtain concerns. However, its role in joint inflammation of rheumatoid arthritis (RA) is largely unknown. Pleckstrin homology domain-containing family O member 1 (PLEKHO1) was previously identified as a negative regulator of osteogenic lineage activity. Here we demonstrated that PLEKHO1 was highly expressed in osteoblasts of articular specimens from RA patients and inflammatory arthritis mice. Genetic deletion of osteoblastic Plekho1 ameliorated joint inflammation in mice with collagen-induced arthritis (CIA) and K/BxN serum-transfer arthritis (STA), whereas overexpressing Plekho1 only within osteoblasts in CIA and STA mice demonstrated exacerbated local inflammation. Further in vitro studies indicated that PLEKHO1 was required for TRAF2-mediated RIP1 ubiquitination to activate NF-kB for inducing inflammatory cytokines production in osteoblasts. Moreover, osteoblastic PLEKHO1 inhibition improved joint inflammation and attenuated bone formation reduction in CIA mice and non-human primate arthritis model. These data strongly suggest that highly expressed PLEKHO1 in osteoblast mediates joint inflammation in RA. Targeting osteoblastic PLEKHO1 may exert dual therapeutic action of alleviating joint inflammation and promoting bone formation in RA.

cell biology↗

bioRxiv2017

Prediction of RNA-protein sequence and structure binding preferences using deep convolutional and recurrent neural networks

RNA regulation is significantly dependent on its binding protein partner, which is known as the RNA-binding proteins (RBPs). Unfortunately, the binding preferences for most RBPs are still not well characterized, especially on the structure point of view. Informative signals hiding and interdependencies between sequence and structure specificities are two challenging problems for both predicting RBP binding sites and accurate sequence and structure motifs mining.\n\nIn this study, we propose a deep learning-based method, iDeepS, to simultaneously identify the binding sequence and structure motifs from RNA sequences using convolutional neural networks (CNNs) and a bidirectional long short term memory network (BLSTM). We first perform one-hot encoding for both the sequence and predicted secondary structure, which are appropriate for subsequent convolution operations. To reveal the hidden binding knowledge from the observations, the CNNs are applied to learn the abstract motif features. Considering the close relationship between sequences and predicted structures, we use the BLSTM to capture the long range dependencies between binding sequence and structure motifs identified by the CNNs. Finally, the learned weighted representations are fed into a classification layer to predict the RBP binding sites. We evaluated iDeepS on verified RBP binding sites derived from large-scale representative CLIP-seq datasets, and the results demonstrate that iDeepS can reliably predict the RBP binding sites on RNAs, and outperforms the state-of-the-art methods. An important advantage is that iDeepS is able to automatically extract both binding sequence and structure motifs, which will improve our transparent understanding of the mechanisms of binding specificities of RBPs. iDeepS is available at https://github.com/xypan1232/iDeepS.

bioinformatics↗

bioRxiv2017

Attention is required for knowledge-based sequential grouping of syllables into words

How the brain sequentially groups sensory events into temporal chunks and how this process is modulated by attention are fundamental questions in cognitive neuroscience. Sequential grouping includes bottom-up primitive grouping and top-down knowledge-based grouping. In speech perception, grouping acoustic features into syllables can rely on bottom-up acoustic continuity cues but grouping syllables into words critically relies on the listeners lexical knowledge. This study investigates whether top-down attention is required to apply lexical knowledge to group syllables into words, by concurrently monitoring neural entrainment to syllables and words using electroencephalography (EEG). When attention is directed to a competing speech stream or cross-modally to a silent movie, neural entrainment to syllables is weakened but neural entrainment to words largely diminishes. These results strongly suggest that knowledge-based grouping of syllables into words requires top-down attention and is a bottleneck for the neural processing of unattended speech.

neuroscience↗

bioRxiv2017

Malthusian Catastrophe: Species Extinction Caused By Oversized Population

There is one pseudo-extinction debt and four occurring conditions for real extinction debt. Since small and oversized populations have a high extinction risk, Pan threshold (upper limit) was calculated for Verhulst-Pear \"logistic\" growth model and logistic model with the Allee effect, an important parameter corresponding to Allee threshold (lower limit).

ecology↗

bioRxiv2016

RNA-protein binding motifs mining with a new hybrid deep learning based cross-domain knowledge integration approach

BackgroundRNAs play key roles in cells through the interactions with proteins known as the RNA-binding proteins (RBP) and their binding motifs enable crucial understanding of the post-transcriptional regulation of RNAs. How the RBPs correctly recognize the target RNAs and why they bind specific positions is still far from clear. Machine learning-based algorithms are widely acknowledged to be capable of speeding up this process. Although many automatic tools have been developed to predict the RNA-protein binding sites from the rapidly growing multi-resource data, e.g. sequence, structure, their domain specific features and formats have posed significant computational challenges. One of current difficulties is that the cross-source shared common knowledge is at a higher abstraction level beyond the observed data, resulting in a low efficiency of direct integration of observed data across domains. The other difficulty is how to interpret the prediction results. Existing approaches tend to terminate after outputting the potential discrete binding sites on the sequences, but how to assemble them into the meaningful binding motifs is a topic worth of further investigation.\n\nResultsIn viewing of these challenges, we propose a deep learning-based framework (iDeep) by using a novel hybrid convolutional neural network and deep belief network to predict the RBP interaction sites and motifs on RNAs. This new protocol is featured by transforming the original observed data into a high-level abstraction feature space using multiple layers of learning blocks, where the shared representations across different domains are integrated. To validate our iDeep method, we performed experiments on 31 large-scale CLIP-seq datasets, and our results show that by integrating multiple sources of data, the average AUC can be improved by 8% compared to the best single-source-based predictor; and through cross-domain knowledge integration at an abstraction level, it outperforms the state-of-the-art predictors by 6%. Besides the overall enhanced prediction performance, the convolutional neural network module embedded in iDeep is also able to automatically capture the interpretable binding motifs for RBPs. Large-scale experiments demonstrate that these mined binding motifs agree well with the experimentally verified results, suggesting iDeep is a promising approach in the real-world applications.\n\nConclusionThe iDeep framework not only can achieve promising performance than the state-of-the-art predictors, but also easily capture interpretable binding motifs. iDeep is available at http://www.csbio.sjtu.edu.cn/bioinf/iDeep

bioinformatics↗