SeqBuster, a bioinformatic tool for the processing and analysis of small RNAs datasets, reveals ubiquitous miRNA modifications in human embryonic cells - PubMed (original) (raw)
SeqBuster, a bioinformatic tool for the processing and analysis of small RNAs datasets, reveals ubiquitous miRNA modifications in human embryonic cells
Lorena Pantano et al. Nucleic Acids Res. 2010 Mar.
Abstract
High-throughput sequencing technologies enable direct approaches to catalog and analyze snapshots of the total small RNA content of living cells. Characterization of high-throughput sequencing data requires bioinformatic tools offering a wide perspective of the small RNA transcriptome. Here we present SeqBuster, a highly versatile and reliable web-based toolkit to process and analyze large-scale small RNA datasets. The high flexibility of this tool is illustrated by the multiple choices offered in the pre-analysis for mapping purposes and in the different analysis modules for data manipulation. To overcome the storage capacity limitations of the web-based tool, SeqBuster offers a stand-alone version that permits the annotation against any custom database. SeqBuster integrates multiple analyses modules in a unique platform and constitutes the first bioinformatic tool offering a deep characterization of miRNA variants (isomiRs). The application of SeqBuster to small-RNA datasets of human embryonic stem cells revealed that most miRNAs present different types of isomiRs, some of them being associated to stem cell differentiation. The exhaustive description of the isomiRs provided by SeqBuster could help to identify miRNA-variants that are relevant in physiological and pathological processes. SeqBuster is available at http://estivill\_lab.crg.es/seqbuster.
Figures
Figure 1.
(A) Workflow of SeqBuster pipeline showing the architecture and connection of pre-analysis and analysis modules. In the pre-analysis module, raw data are processed for recognition and annotation. Annotation can be performed through the web server that offers the miRNA and miRNA-precursor databases or through a stand-alone version using any custom database. The processed and annotated data are stored in a MySQL database. The web interface permits the analysis of the data using several R-based packages. The output of every analysis is visualized through a Dynamic HTML format and stored in the server or downloaded to the local machine. (B) Scheme of the main menu at SeqBuster home page. The different choices offered by each option in the menu are highlighted in light yellow boxes.
Figure 2.
Percentage of reads with a mismatch at different positions of the reference miRNA detected by SeqBuster, considering two different annotation strategies. Penalty and reward parameters of −3 and 1 (black bars) or −2 and 3 (grey bars) were used. In both strategies a word size of 7 was considered.
Figure 3.
The ‘IsomiR distribution’ package scheme. (A) Within the ‘IsomiR analysis’ several packages appear in a general menu. After selecting the ‘isomiR distribution’ package the samples for the analysis should be chosen. Up to four samples can be loaded to the analysis. Different options and parameters may be configured in order to customize the study (tutorial 4). (B) In the output analysis, a histogram displays the proportion of miRNAs with different types of isomiRs in all the selected samples. For every type of variability, the upper part of the graph shows the proportion of miRNAs presenting one (white), two (grey) or more than two (black) isomiRs. The abundance of the isomiR with respect to the corresponding reference miRNA is mirrored in the lower graph in a brown color scale. The five brown color intensities from dark to light indicate the frequency of the isomiR with respect to that of the reference miRNA: 1, > 80%; 2, 60–80%; 3, 40–60%; 4, 20–40% and 5, < 20%. Below the graph, a table helps to obtain the complete information of the analysis. All the miRNAs contained in the histogram can be listed by clicking on the corresponding link. Those miRNAs highlighted in pink are commonly detected in all the samples examined.
Figure 4.
Summary of the main packages available in SeqBuster IsomiR analysis module. For the three packages, the parameters that can be configured (tutorial 4) and the analysis output are shown. (A) ‘IsomiRs by nucleotide position’ package. This example shows 5′-trimming variants involving three positions upstream and three positions downstream of the reference-miRNAs. The output shows the proportion of miRNAs with a trimming variant in a specific position. In the upper bars, the color pattern indicates the proportion of miRNAs showing a specific nucleotide being involved in the trimming variants. The lower bars show the proportion of the isomiRs with respect to the corresponding reference miRNA in a brown color scale, as described in Figure 3. A list of miRNAs involved in each type of variant is displayed when clicking inside the table. (B) ‘IsomiR full description’ package. The output analysis of this example shows a list of some of the miRNAs presenting 5′-trimming variants involving three positions upstream and three positions downstream of the reference-miRNAs. Every cell represents a position. The color pattern on the left half of the cell indicates the type of nucleotide present in the isomiR, and that of the right side the proportion of the isomiR with respect the reference miRNA. (C) ‘Nt-substitution pattern’ package. The output analysis of this example shows a table with the number of miRNAs presenting any of the 12 possible nt-substitution events in positions 2–5 of the reference miRNA. A summary table showing the number of miRNA with any of the 12 possible nt-changes significant changes is also represented. The overall nt-substitution pattern occurring in a statistically significant number of miRNAs is shown in another table. The list of miRNAs for each class of nt-substitution can be retrieved by clicking on the summary table.
References
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Miscellaneous