GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM4146764

Query DataSets for GSM4146764

Status

Public on Apr 24, 2020

Title

WT_m2_d17 (Amplicon Bisulfite)

Sample type

SRA

Source name

Wild-type mESCs

Organism

Mus musculus

Characteristics

strain: C57BL6/J
tissue: mESCs
genotype/variation: Wild-type TET
genotype/variation: Dnmt3a/3b intact
timepoint: day 17
replicate: replicate 2

Extracted molecule

genomic DNA

Extraction protocol

Approximately 2ug of extracted genomic DNA from days 0, 4, 8, 10, 13, 17 and 29 were first mixed with 3.2pg of both unmethylated Lambda bacteriophage and in vitro methylated T7 bacteriophage DNA. Addition of the bacteriophage DNA was used to control for bisulfite conversion efficiency. Samples were then bisulfite converted using the EpiTect kit from Qiagen per the manufacturer’s instructions. Primers designed to amplify UMRs, LMR and FMR regions (88 in total, Supplemental Table 1) were distributed in 96 well plates and amplification was carried out using Amplitaq gold using the following thermocycler settings (all temperatures are in celcius and all incubation times are 30” unless specified): 1 cycle of 95⁰ for 9’, 20 cycles of touchdown with 95⁰ melt, 55⁰-51⁰ annealing, and extension at 72⁰, followed by 36 cycles of 95⁰ melt, 51⁰ annealing and 72⁰ extension. The amplicon reactions were then mixed, run out on an agarose gel and DNA extracted. Libraries were then constructed using the NEBNext ChIP-seq Library Prep kit (#E6240) as per the manufacturers instructions, indexed, pooled and sequenced using the Illumina MiSeq platform in 250bp paired-end mode.

Library strategy

Bisulfite-Seq

Library source

genomic

Library selection

RANDOM

Instrument model

Illumina MiSeq

Description

Mock-Cre transduction day 17

Data processing

Alignment: For amplicon bisulfite samples, the last 100 base pairs were trimmed due to reduced sequencing quality using the preprocessReads function from the Bioconductor QuasR package with the truncateEndBases = 100 parameter. Amplicon sequences were then aligned to the mm9 build using Rbowtie in the QuasR package and the following parameters: genome="BSgenome.Mmusculus.UCSC.mm9", paired="fr", bisulfite="undir". For SureSelect samples, reads were aligned to the mm9 build using Rbowtie in the QuasR environment as for amplicon sequencing above except with the parameter bisulfie = "dir".
Methylation level determination: Methylation levels for CpGs in amplicons was determined using the AmpliconViews function from the R package AmpliconBiSeq with the parameters ‘conv = 80’ and exp.var = ‘90’. Methylation levels for SureSelect samples was calculated using the qMeth function in the QuasR package. For the SureSelect methylation data (Supplemental_Table4), read counts for replicates of the same time points were added together before determining methylation levels. Coverage columns for Supplemental_Table4 reflect summed counts for the triplicate experiments.
Filtering: A minimum of 100x coverage was required for all methylation calls in the amplicon bisulfite data. If coverage did not reach this threshold for any time point, the respective methylation level was set to NA. We further removed all CpGs with NA at day 0 or with more than one NA during the time course across all replicates. In addition, we filtered out one amplicon (chr7:149767665-149767784) containing 9 CpGs due to high variability in methylation calls. From a total of 588 CpGs initially quantified, this filtering procedure resulted in 405 CpGs. For SureSelect, a minimum of 50x coverage in all time points and replicates was used to filter CpGs for further analysis.
Rate determination: For methylation and demethylation rate determination, please see text.
Genome_build: mm9
Supplementary_files_format_and_content: Table files with CpG coordinates, methylation level, coverage, rate of methylation, rate of demethylation and tet activity (SureSelect only). For sample identication in Supplemental tables, there are 4 elements in the sample name that reflect the sample composition. Taking the sample name 'TTKO_cre1_d4_methylation' as an example, the TTKO refers to 'Tet Triple Knock-Out', while the 'cre1' refers to 'Cre transduction, replicate 1', the element 'd4' refers to 'Day 4 post transduction' and finally 'methylation' is the percent methylated alleles at that CpG. Other acronyms used are 'WT' for 'Wild-type', 'coverage' for number of reads covering that particular CpG in the sample, and 'mock' refers to the negative control in cre transduction. More specifically, 'm' refers to 'mock' and indicates addition of DMSO to the cells instead of Cre protein. The column 'MLE_kMe' reflects the rate of methylation while 'MLE_kDe' reflects the rate of demethylation (described in detail in the manuscript). The confidence column reflects the confidence level of rate inferrence. Lastly, the 'tag' column in Supplemental Table 2 and 3 reflects whether the CpG resides in an LMR (Low Methylated Region), FMR (Fully Methylated Region), UMR (Unmethylated Region), metIsland (Methylated CpG Island), NP-specific (Neural Progenitor specific) and ES-specific (Embyonic Stem cell specific) region. The designation 'Constitutive' reflects that the CpG resides in regions with the methylation state of interest in different cell types. In Supplemental Table 2, 'inferredInWTandTET' indicates rates inferred in wild-type and TTKO timecourse, while 'kDeWT_predicted' indicates predicted demethylation rates in wild-type cells using TTKO time course information and wild-type steady state measurements.

Submission date

Nov 01, 2019

Last update date

Apr 24, 2020

Contact name

Dirk Schuebeler

Organization name

Friedrich Miescher Institute for Biomedical Research

Street address

Maulbeerstrasse 66

City

Basel