GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM995320

Query DataSets for GSM995320

Status

Public on Oct 01, 2012

Title

main_oan_bra

Sample type

SRA

Source name

Brain

Organism

Ornithorhynchus anatinus

Characteristics

tissue/cell: Brain
age: Adult
gender: Male

Extracted molecule

total RNA

Extraction protocol

We extracted high-quality RNA from the 38 samples described above using standard protocols. We then prepared small RNA-seq libraries for each of these samples using Illumina Small RNA v1.5 Sample Preparation protocol with the following optimizations. We first purified small RNAs (18-30 nucleotides) from total RNA using denaturing PAGE gel electrophoresis. We used 10% Novex TBE PAGE gel instead of the 6% described in the protocol to ensure a better separation of the cDNA constructs during the purification step.

Library strategy

RNA-Seq

Library source

transcriptomic

Library selection

size fractionation

Instrument model

Illumina Genome Analyzer IIx

Data processing

Base calling was performed using the standard Illumina software used on the Illumina Genome Analyzer Iix. Quality score in the raw fastq files are encoded using Phred 64.
Read were trimmed using a standard in-house procedure, removing the following adpater sequence: ATCTCGTATGCCGTCTTCTGCTTG
Reads of length 15-23 nucleotides were then selected to perform miRNA detection (see Methods in the related paper) and then to estimate expression level of detected miRNAs. To compute expression levels, we mapped 15-23 nt reads on the miRNA mature sequences (+/- 3base pair flanks) using bowtie (allowing for multiple mappings but no mismatch: bowtie -v 0 -a. ). We also mapped the reads on their respective genome, masking the miRNA precursor sequences, to identify reads mapping to genomic location other than miRNAs and excluded from the dataset miRNAs for which more than 10% of the mapped reads mapped also to other genomic locations.
For each mature miRNA, reads were then counted. When two (or more) miRNA genes shared more than 10% of their reads, a single joint expression value was computed for them, thus only counting once every single read (i.e., the two miRNAs are considered as one in the relevant analyses). If two (or more) miRNA genes shared less than 10% of their reads, then shared reads were equally distributed across the miRNA genes. Thus, in all cases, multiple mapping reads are never counted more than once for expression values.
Normalization across tissues was performed using edgeR (Robinson et al. 2010), which takes into account variable library sizes and corrects for biases in expression level estimates caused by highly expressed genes in a subset of tissues.
Genome_build: Ensembl Release 57
Supplementary_files_format_and_content: For each species, a processed data file is available containing a normalized expression matrix (mir.id denoted the miRNA gene ID as used in Supplementary Table 1).

Submission date

Aug 30, 2012

Last update date

May 15, 2019

Contact name

Henrik Kaessmann

Organization name

University of Lausanne

Department

Center For Integrative Genomics

Lab

Kaessmann