Mining SNPs from EST sequences using filters and ensemble classifiers

AcrPred: A hybrid optimization with enumerated machine learning algorithm to predict Anti-CRISPR proteins

CRISPR-Cas, as a tool for gene editing, has received extensive attention in recent years. Anti-CRISPR (Acr) proteins can inactivate the CRISPR-Cas defense system during interference phase, and can be used as a potential tool for the regulation of gene editing. In-depth study of Anti-CRISPR proteins is of great significance for the implementation of gene editing. In this study, we developed a high-accuracy prediction model based on two-step model fusion strategy, called AcrPred, which could produce an AUC of 0.952 with independent dataset validation. To further validate the proposed model, we compared with published tools and correctly identified 9 of 10 new Acr proteins, indicating the strong generalization ability of our model. Finally, for the convenience of related wet-experimental researchers, a user-friendly web-server AcrPred (Anti-CRISPR proteins Prediction) was established at http://lin-group.cn/server/AcrPred, by which users can easily identify potential Anti-CRISPR proteins.

Survey of Natural Language Processing Techniques in Bioinformatics

Informatics methods, such as text mining and natural language processing, are always involved in bioinformatics research. In this study, we discuss text mining and natural language processing methods in bioinformatics from two perspectives. First, we aim to search for knowledge on biology, retrieve references using text mining methods, and reconstruct databases. For example, protein-protein interactions and gene-disease relationship can be mined from PubMed. Then, we analyze the applications of text mining and natural language processing techniques in bioinformatics, including predicting protein structure and function, detecting noncoding RNA. Finally, numerous methods and applications, as well as their contributions to bioinformatics, are discussed for future use by text mining and natural language processing researchers.

Human cytochrome P450 and personalized medicine

Personalized medicine has become a hot topic ascribed to the development of Human Genome Project. And currently, bioinformatics methodology plays an essential role in personal drug design. Here in this review we mainly focused on the basic introduction of the SNPs of human drug metabolic enzymes and their relationships with personalized medicine. Some common bioinformatics analysis methods and latest progresses and applications in personal drug design have also been discussed. Thus bioinformatics studies on SNPs of human CYP450 genes will contribute to indicate the most possible genes that are associated with human diseases and relevant therapeutic targets, identify and predict the drug efficacy and adverse drug response, investigate individual gene specific properties and then provide personalized and optimal clinic therapies.

Transcriptome analysis of the differences in gene expression between testis and ovary in green mud crab (Scylla paramamosain)

Background

The green mud crab (Scylla paramamosain) is the most prevalent crustacean on the southeast coast of China. The molecular regulatory mechanism of sex determination and gonadal differentiation in this species has received considerable attention in recent years because of the huge differences—both biological and economic—between male and female crabs. In this study, next-generation sequencing technology was used to develop deep-coverage transcriptomic sequencing data for the testis and ovary of S. paramamosain.

Results

A total of 365,116 reads (testis 171,962, ovary 193,154) with an average sequence length of 285 bp were produced from testis and ovary cDNA libraries. After filtering out contaminating reads, the clean reads were assembled, producing a total of 21,791 isotigs and leaving 22,814 reads as singlets. Using the BLASTX program, 3,471 unique sequences (2,275 isotigs and 1,196 singletons) were annotated with known protein sequences from the NCBI non-redundant (Nr) protein sequence database. The Gene Ontology and KEGG (Kyoto Encyclopedia of Genes and Genomes) analyses allowed the 224 unique sequences that were annotated with enzyme code (EC) numbers to be mapped into 174 KEGG pathways. After comparing the ovary and testis libraries, 4,021 gonad-differentially, 10,522 ovary-specifically, and 19,013 testis-specifically expressed genes were identified. Moreover, 33 ovary-specific, 14 testis-specific, and 34 gonad-differential transcripts were confirmed by semi-quantitative PCR and quantitative real-time PCR. In addition, 8,610 putative simple sequence repeats (SSRs) and 23,879 potential single nucleotide polymorphisms (SNPs) were identified.

Conclusion

This is the first large-scale RNA sequencing of S. paramamosain to be reported. We have identified many important functional genes and made a preliminary attempt to construct the regulatory network involved in the gonadal development of crustaceans. The annotated transcriptome data will provide fundamental support for future research into the reproduction biology of S. paramamosain. A large number of candidate SSRs and SNPs were detected, which could be used as genetic markers for population genetics and functional genomics in this species.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-585) contains supplementary material, which is available to authorized users.

Effective sample selection for classification of pre-miRNAs

To solve the class imbalance problem in the classification of pre-miRNAs with the ab initio method, we developed a novel sample selection method according to the characteristics of pre-miRNAs. Real/pseudo pre-miRNAs are clustered based on their stem similarity and their distribution in high dimensional sample space, respectively. The training samples are selected according to the sample density of each cluster. Experimental results are validated by the cross-validation and other testing datasets composed of human real/pseudo pre-miRNAs. When compared with the previous method, microPred, our classifier miRNAPred is nearly 12% more accurate. The selected training samples also could be used to train other SVM classifiers, such as triplet-SVM, MiPred, miPred, and microPred, to improve their classification performance. The sample selection algorithm is useful for constructing a more efficient classifier for the classification of real pre-miRNAs and pseudo hairpin sequences.