Computational identification of vesicular transport proteins from sequences using deep gated recurrent units architecture

Le, Nguyen Quoc Khanh; Yapp, Edward Kien Yee; Nagasundaram, Nagarajan; Chua, Matthew Chin Heng; Yeh, Hui‐Yuan

doi:10.1016/j.csbj.2019.09.005

Cited by 45 publications

(40 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Database for Annotation, Visualization, and Integrated Discovery (DAVID, 6.8 version, http://david.ncifcrf.gov) and Kyoto Encyclopedia of Genes and Genomes (KEGG, https://www.genome.jp/kegg/) were adopted to extract biological information about the DEGs (Huang et al, 2007). Gene ontology (GO) is widely used in biological research (Le et al, 2019a;Le, Yapp & Yeh, 2019b).…”

Section: Kegg and Go Enrichment Analyses Of Degsmentioning

confidence: 99%

CKS2 and RMI2 are two prognostic biomarkers of lung adenocarcinoma

Xiao

Dong

Yang

et al. 2020

PeerJ

View full text Add to dashboard Cite

Background Lung adenocarcinoma (ACA) is the most common subtype of non-small-cell lung cancer. About 70%–80% patients are diagnosed at an advanced stage; therefore, the survival rate is poor. It is urgent to discover accurate markers that can differentiate the late stages of lung ACA from the early stages. With the development of biochips, researchers are able to efficiently screen large amounts of biological analytes for multiple purposes. Methods Our team downloaded GSE75037 and GSE32863 from the Gene Expression Omnibus (GEO) database. Next, we utilized GEO’s online tool, GEO2R, to analyze the differentially expressed genes (DEGs) between stage I and stage II–IV lung ACA. The using the Cytoscape software was used to analyze the DEGs and the protein-protein interaction (PPI) network was further constructed. The function of the DEGs were further analyzed by cBioPortal and Gene Expression Profiling Interactive Analysis (GEPIA) online tools. We validated these results in 72 pairs human samples. Results We identified 109 co-DEGs, most of which were involved in either proliferation, S phase of mitotic cell cycle, regulation of exit from mitosis, DNA replication initiation, DNA replication, and chromosome segregation. Utilizing cBioPortal and University of California Santa Cruz databases, we further confirmed 35 hub genes. Two of these genes, encoding CDC28 protein kinase regulatory subunit 2 (CKS2) and RecQ-mediated genome instability 2 (RMI2), were upregulated in lung ACA compared with adjacent normal tissues. The Kaplan–Meier curves revealed upregulation of CKS2 and RMI2 are associated with worse survival. Using CMap analysis, we discovered 10 small molecular compounds that reversed the altered DEGs, the top five are phenoxybenzamine, adiphenine, resveratrol, and trifluoperazine. We also evaluated 72 pairs resected samples, results revealed that upregulation of CKS2 and RMI2 in lung ACA were associated with larger tumor size. Our results allow the deeper recognizing of the mechanisms of the progression of lung ACA, and may indicate potential therapeutic strategies for the therapy of lung ACA.

show abstract

Section: Kegg and Go Enrichment Analyses Of Degsmentioning

confidence: 99%

CKS2 and RMI2 are two prognostic biomarkers of lung adenocarcinoma

Xiao

Dong

Yang

et al. 2020

PeerJ

View full text Add to dashboard Cite

show abstract

“…To understand the generalization performance, CNA_origin was also tested by independent datasets. In this work, the precision (P), recall (R), accuracy (ACC), and F1-score were adopted to assess the performance of the corresponding method; they have been used as measurement metrics in previous works (Le et al, 2018 , 2019a ). They are defined as Equation (6).…”

Section: Resultsmentioning

confidence: 99%

A Deep Learning Framework to Predict Tumor Tissue-of-Origin Based on Copy Number Alteration

Liang

Wang

Yang

et al. 2020

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Cancer of unknown primary site (CUPS) is a type of metastatic tumor for which the sites of tumor origin cannot be determined. Precise diagnosis of the tissue origin for metastatic CUPS is crucial for developing treatment schemes to improve patient prognosis. Recently, there have been many studies using various cancer biomarkers to predict the tissue-of-origin (TOO) of CUPS. However, only a very few of them use copy number alteration (CNA) to trance TOO. In this paper, a two-step computational framework called CNA_origin is introduced to predict the tissue-of-origin of a tumor from its gene CNA levels. CNA_origin set up an intellectual deep-learning network mainly composed of an autoencoder and a convolution neural network (CNN). Based on real datasets released from the public database, CNA_origin had an overall accuracy of 83.81% on 10-fold cross-validation and 79% on independent datasets for predicting tumor origin, which improved the accuracy by 7.75 and 9.72% compared with the method published in a previous paper. Our results suggested that the autoencoder model can extract key characteristics of CNA and that the CNN classifier model developed in this study can predict the origin of tumors robustly and effectively. CNA_origin was written in Python and can be downloaded from https://github.com/YingLianghnu/CNA_origin.

show abstract

“…This suggests that homopeptides may be selected against in structured interaction interfaces, perhaps since they introduce a lack of interaction specificity, or increase the likelihood of off-target binding. Also, the gene ontology results indicate that homopeptide occurrences may be useful information for the discrimination of protein function from the analysis of sequences (Huntley et al 2014;Huntley et al 2015;Jiang et al 2016;Le et al 2019a;Le et al 2019b;Mutowo-Meullenet et al 2013).…”

Section: Gene Ontology Enrichmentsmentioning

confidence: 99%

Peer Review #1 of "The relationship between protein domains and homopeptides in the Plasmodium falciparum proteome (v0.1)"

2020

View full text Add to dashboard Cite

The proteome of the malaria parasite Plasmodium falciparum is notable for the pervasive occurrence of homopeptides or low-complexity regions (i.e., regions that are made from a small subset of amino-acid residue types). The most prevalent of these are made from residues encoded by adenine/thymidine (AT)-rich codons, in particular asparagine. We examined homopeptide occurrences within protein domains in P. falciparum. Homopeptide enrichments occur for hydrophobic (e.g., valine), or small residues (alanine or glycine) in short spans (<5 residues), but these enrichments disappear for longer lengths. We observe that short asparagine homopeptides (<10 residues long) have a dramatic relative depletion inside protein domains, indicating some selective constraint to keep them from forming. We surmise that this is possibly linked to co-translational protein folding, although there are specific protein domains that are enriched in longer asparagine homopeptides (≥10 residues) indicating a functional linkage for specific poly-asparagine tracts. Top gene ontology functional category enrichments for homopeptides associated with diverse protein domains include 'vesicle-mediated transport', and 'DNA-directed 5'-3' RNA polymerase activity', with various categories linked to 'binding' evidencing significant homopeptide depletions. Also, in general homopeptides are substantially enriched in the parts of protein domains that are near/in IDRs. The implications of these findings are discussed.

show abstract

Computational identification of vesicular transport proteins from sequences using deep gated recurrent units architecture

Cited by 45 publications

References 46 publications

CKS2 and RMI2 are two prognostic biomarkers of lung adenocarcinoma

CKS2 and RMI2 are two prognostic biomarkers of lung adenocarcinoma

A Deep Learning Framework to Predict Tumor Tissue-of-Origin Based on Copy Number Alteration

Peer Review #1 of "The relationship between protein domains and homopeptides in the Plasmodium falciparum proteome (v0.1)"

Contact Info

Product

Resources

About