GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM607093

Query DataSets for GSM607093

Status

Public on Nov 07, 2010

Title

barcode10 (sample not used)

Sample type

SRA

Source name

stem cell culture

Organism

Mus musculus

Characteristics

strain: KH2
cell type: embryonic stem cells
developmental stage: UNDIFF

Treatment protocol

The "nuclease" samples were partially digested with P1 ssRNA-specific endonuclease. The "PNK" samples were treated with polynucleotide kinase.
Digestion was carried out using P1, which displays robust activity at high temperatures (~70°C) at acidic pH (~4.5) with little sequence specificity. We performed digests at buffer conditions close to physiological conditions and at 37°C, where P1 activity is less robust. All of the following steps were performed with two different input RNA samples in parallel: nuclear RNA from mouse undifferentiated embryonic stem cells (KH2) or the KH2 line differentiated into neural precursor cells. The two different nuclear RNA samples (500 ng) were suspended in a nuclease digestion buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl2, 5 mM MgCl2 and 0.01 mM Zn(OAc)2) at 1 ng/ul and denatured at 55°C for 5 min and then placed at 37°C for 10 min to facilitate refolding. Parallel reactions with no nuclease or P1 (Sigma) were then initiated from this master mix. For nuclease digestion, 1 ug of enzyme was added, reactions were incubated for 1 h at 37°C and then stopped by addition of 10 mM EDTA and acid-phenol-chloroform extraction. For the no-nuclease conditions, two samples (control and PNK treatment) were used in which half the material was treated with T4 PNK (30 U; USB) for 1 h in a buffer that favors both the 5' phosphorylation activity and the 3' phosphatase activity (50 mM imidazole-HCl, pH 6.0, 10 mM MgCl2 and 10 mM DTT).
RNA fragments (from each of "nuclease", "control", or "PNK" treatments) were size-selected to 20-100 bases, then cloned using the SOLiD Small RNA Expression Kit protocol (https://www3.appliedbiosystems.com/cms/groups/mcb_marketing/documents/generaldocuments/cms_054973.pdf) and PCR-amplified. Equimolar quantities of each barcoded sample were mixed for the sequencing run.

Growth protocol

KH2 mouse embryonic stem cells (Open Biosystems) were cultured under standard conditions as in Beard et al (Genesis 2006), but on gelatin-coated dishes rather than feeder layers. UNDIFF CELLS: For the undifferentiated samples, cells were grown with leukemia inhibitory factor (LIF) in GMEM medium (Skarnes et al, Methods in Enzymology 2000). D5NP CELLS: To generate neural precursor cells, the KH2 cells were replated on fresh gelatin-coated dishes at a density of 1.5 x 104 cells/cm2 and the medium was switched to N2/B27 18 hours later (Ying et al, Nature Biotechnology 2003). The cultures were grown for 5 days under neural differentiation conditions before harvesting RNA.

Extracted molecule

nuclear RNA

Extraction protocol

Adherent cells were washed twice with PBS and then mechanically dissociated from the dishes with PBS with 1 mM EDTA at 4°C. After pelleting at 500g, the cells were resuspended in lysis buffer minus NP40 (10 mM Tris-HCl pH 7.5, 75 mM NaCl and 1 mM MgCl2) and repelleted. The cells were then gently resuspended in the lysis buffer (same buffer above but with 0.3% NP40, mammalian protease inhibitor cocktail (Sigma) and Superase-In RNase inhibitor (Ambion, 200 U/ml)). After a 5-min incubation on ice, the lysates were spun at 900g for 5 min. The resulting supernatant was removed and transferred to TRIzol reagent (Invitrogen) for cytoplasmic RNA isolation. The nuclear pellet was gently resuspended and pelleted twice in lysis buffer to wash away contaminating cytoplasmic components. The resuspended nuclei were purified by sedimentation through a 25% sucrose cushion (dissolved in lysis buffer without NP40) spun at 16,000g for 15 min. Under these sedimentation conditions, the nuclei form a pellet. The resulting nuclei were disrupted by sonication in TRIzol, and the RNA was isolated by the standard reagent protocol, and treated with DNase I (Ambion) and acid-phenol-chloroform extraction to remove genomic DNA. The effectiveness of the cytoplasmic or nuclear enrichment strategy was qualitatively determined by reverse-transcription PCR against U2 or U6 snRNA (nuclear RNA-enriched) and spliced B-actin mRNA (cytoplasmic RNA-enriched) in the two samples. Nuclear RNA integrity was assayed by [32P]pCp 3' end labeling with pCp and RNA ligase (NEB) followed by 10% urea PAGE analysis to visualize the characteristic nuclear RNA mobilities of U1, U2 and U3 RNAs and tRNA.

Library strategy

ncRNA-Seq

Library source

transcriptomic

Library selection

size fractionation

Instrument model

AB SOLiD System 3.0

Description

nuclear RNA
library strategy: BARCODE
instrument model: ABI SOLiD3

Data processing

Raw sequencing colorspace reads from the SOLiD3 run (10 barcodes) were split up by barcode using custom software to generate the 10 colospace FASTA files provided with this GEO submission. The same was done for the QUAL files in this submission.
We used the ABI Small RNA Analysis Pipeline (http://solidsoftwaretools.com), which takes as input sequencing reads in colorspace, filters out reads mapping to undesirable sequences, optionally maps unfiltered reads to a set of known canonical (small) RNA sequences, and maps the remaining unmapped reads to the genome. Our filtering database contained ABI SOLiD sequencing adapters and ribosomal RNA, including pre-rRNA (internally and externally transcribed spacer sequences); reads mapping to this database were removed. The remaining reads were mapped directly to the mm9 mouse genome assembly converted to SOLiD colorspace. The read sequence was inferred after the mapping process; that is, each colorspace read was converted to a sequence of bases that best satisfied the colorspace read as a match to a colorspace version of the reference genome. This accommodates the possibility that the reference genome sequence may be different from the RNA sequence. The resulting base-space sequences of every mapped read were output in the fourth column of a 6-column BED-format file, which are the "processed data" files in this GEO submission.
When mapping to the genome, the pipeline first creates "seed" mappings of the first 30 positions of each read (corresponding to the 5'-most bases in the cloned RNA fragment), allowing up to two colorspace mismatches in those regions. To determine how much, if any, of the remainder of the 50mer read sequence consists of the adapter (as opposed to RNA fragment), a set of local alignments of the entire read is made to the reference (at the seed mapping position) followed by each possible concatenation of a subset of adapter bases. The mapping among these alignments with the lowest number of mismatches determines the "extended" read length, excluding adapter sequence. The total number of color mismatches allowed in this extension step was six, to allow for the decay in accuracy towards the 3' end of SOLiD reads.
Note that we allowed each read to map to the genome multiple times, up to 15 loci per chromosome (each chromosome mapping was run as an independent computing job). A post-processing step (using custom code) summed those counts to determine the number of genome-wide read mappings for each read. These counts are in the suffix of each read name (4th column of BED file in the processed files in this GEO submission).

Submission date

Oct 11, 2010

Last update date

May 15, 2019

Contact name

Andrew V Uzilov

E-mail(s)

auzilov@ucsc.edu

Organization name

University of California, Santa Cruz

Street address

1156 High St, MS SOE3