sRNAde

Consensus differential expression (DESeq, edgeR and NOISeq), visual cluster analysis and sequencing statistics.

image

Description

sRNAde is a module for the detection of differentially expressed sRNA. The input is either a number of sRNAbench output folders or a user-given expression matrix. For the detection of differentially expressed RNAs it applies three widely used methods: edgeR, DESeq and NOISeq. Apart of the output from the individual methods, it provides a consensus differential expression file. Additionally, this tool can perform cluster analysis (heat maps), isomiR analysis and it gives a summary on the sequencing statistic of all used samples (if the input has been from sRNAbench). Result example.

Input

Users should provide a list of sRNAbench jobIDs on which differential expression analysis of miRNAs will be performed. There are two ways to do this:

  • A list of jobIDS and group names: On the "Use Job IDs" tab the user can provide a list of sRNAbench job IDs (comma separated) and group names (comma separated, as many as the user wants to include in the analysis). In a second step, each sample can be assigned to the group it belongs using a dropdown selector (see image below).

image

  • A list of colon and hash jobIDs using # to separate groups (GroupString): job IDs should be provided in the following way:

    f1_1:f2_1#f1_2:f2_2 being f1_1 the ID of the first sample of the first group (controls in a case/control study), f2_1 the second sample of the first group, f1_2 the first sample of the second group, etc. That means that groups are separated by hashes (#), while samples are separated by colon ( : ).

    Note, the ID is assigned by sRNAbench and can be found either in the URL or in the output page

    For example, the test data run can also be launched using this grpString:

    D42PPR3ZVTVQBK8:EYHC83BJX80ATIL:UWCUSRT3IZO3ME2:XXEYZ4EZ55BHKUG:YH5LQ JZVNP9W3NX:5DZDVDGE0AZ5GQ0#51UQ20HBJ3XUWPJ:6IJISG8JIDJC0CA:FKZ2D88FZN WZCS5:OE1ZNRX2RSJ0OV1:OUSQYGT9V0V6ZA5:ZBM4WYEO1C80DTU

    and the following group description string:

    exosome#cell

The test data comprises therefore 2 experimental groups (exosome and cell). sRNAde calculates the differential expression between all possible combinations (only exosome vs cell in this case).

The user can also provide a description of the samples separated by colons. If no description is given, the sRNAbench job name is used by default. Finally, the user can provide differential expression thresholds for NOISeq, DESeq, DEseq2, ttest and edgeR.

Please, note they are all adjusted p-value.

Results

image

  1. Results summary: Mapping statistics of reads. a. Number of mapped or assigned reads b. Percentage of mapped or assigned reads c. Percentage per detected RNA category d. Detected number of miRNAs and precursor sequences

  2. Preprocessing/QC: Statistics of reads preprocessing and quality control. a. Preprocessing statistics b. Read length distribution (full, analysis and genome mapped)

  3. Mapping statistics:The user can visualize the genome mapping distribution.

  4. miRNA and isomiR statistics:The user can visualize the fraction of different isomiR classes per sample.

  5. Differential expression: We provide the result of differential expression of the 5 methods for each comparison between groups defined by user. In this section, only significantly differentially expressed miRNAs are shown (the cut-off value can be provided by the user). The result description of each method is widely explained in their manuals: