Open Access Research article

The prediction of the porcine pre-microRNAs in genome-wide based on support vector machine (SVM) and homology searching

Zhen Wang12, Kan He123, Qishan Wang12, Yumei Yang12 and Yuchun Pan12*

Author Affiliations

1 School of Agriculture and Biology, Department of Animal Science, Shanghai Jiao Tong University, Shanghai, 200240, PR China

2 Shanghai Key Laboratory of Veterinary Biotechnology, Shanghai, 200240, PR China

3 Department of Biology, Faculty of Science, Hong Kong Baptist University, Hong Kong, China

For all author emails, please log on.

BMC Genomics 2012, 13:729  doi:10.1186/1471-2164-13-729

Published: 27 December 2012

Additional files

Additional file 1:

The list of porcine pre-miRNA candidates predicted by SVM-based classifier. The data provided represent the list of porcine pre-miRNA candidates predicted by SVM-based classifier in the whole genome of the pigs, and containing the information of their length, location in chromosome and genome location clusters.

Format: XLSX Size: 183KB Download file

Open Data

Additional file 2:

The result of density analysis of pre-miRNA and QTL in chromosome. The data provided the information of the number of pre-miRNA and QTL and their density in each chromosome.

Format: XLSX Size: 13KB Download file

Open Data

Additional file 3:

The list of porcine known pre-miRNA fragments of 90nt detected by SVM-based classifier. The data provided represents the list of porcine known pre-miRNA detected by SVM-based classifier in the whole genome of the pigs, and containing the information of their length, location in chromosome and the name of the represented known pre-miRNA.

Format: XLSX Size: 27KB Download file

Open Data

Additional file 4:

The list of porcine pre-miRNA candidates predicted by homology searching. The data provided represent the list of porcine pre-miRNA candidates predicted by homology searching, and containing the information of their length and location in chromosome.

Format: XLSX Size: 20KB Download file

Open Data

Additional file 5:

Local sequence-structure features of a hairpin were denoted by the left-triplet coding. Left-triplet elements are used to represent the local structure sequence features of a hairpin. The nucleotide type at the left and three local continuous substructures compose the left-triplet element. The appearances of all 32 possible triplet elements are counted along a hairpin segment to form a 32-dimensional vector, which is normalized to be the input vector for SVM.

Format: JPEG Size: 38KB Download file

Open Data

Additional file 6:

The 46 features used by SVM-based porcine pre-miRNAs classifier.

Format: DOCX Size: 14KB Download file

Open Data

Additional file 7:

The primary sequence of the has-let-7e precursor and the locations of some terms in the secondary structure. The upper part gives the primary structure of has-let-7e and the lower one shows the secondary structure and the correlative terms with varied colors.

Format: JPEG Size: 98KB Download file

Open Data