GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM1263820

Query DataSets for GSM1263820

Status

Public on Jun 30, 2014

Title

Liver_ERa_ChIP_E2_repl2

Sample type

SRA

Source name

liver

Organism

Mus musculus

Characteristics

treatment: 10 nM beta-estradiol (E2)
chip antibody: anti mouse ER-alpha
strain: C57BL/6
tissue: liver

Treatment protocol

Aortas and livers (tip of the major lobe) were treated ex vivo with either 10-8M 17b-estradiol (E2, the active endogenous form of estrogen) or ethanol vehicle for 45 minutes at 37oC in DMEM (Invitrogen).

Growth protocol

Wild-type female C57BL/6 mice were ovariectomized at 6-7 weeks of age. Aortas and livers were harvested from these animals one week post surgery. All animals were handled according to National Institutes of Health standards and protocols approved by the Tufts Medical Center Institutional Animal Care and Use Committee.

Extracted molecule

genomic DNA

Extraction protocol

Samples were cross-linked with 1% formaldehyde at 37°C for 15 minutes and quenched with 0.125M glycine. Tissues were then rinsed with 1X PBS, minced with scissors, and homogenized. Samples were pelleted by brief centrifugation and resuspended in 0.5 ml lysis buffer (1% SDS, 10mM EDTA pH 8.0, 50mM Tris pH 8.0, 1x Complete Protease Inhibitors (Roche 1187358001)). DNA was fragmented by seven rounds of sonication with a Sonifier 250 tip sonicator (Branson Ultransonics) at output setting 5 for 5 seconds followed by 1 minute on ice. Samples were centrifuged at 13500 rpm for 15 minutes at 4°C and the supernatant collected. 14.5 l of Protein A Dynabeads and 14.5 l of protein G Dynabeads (invitrogen) were washed by magnetic pull down 3x at 4oC with PBS+5mg/ml BSA. 15 l of anti-ER HC-20 antibody (Santa Cruz) and 10 l of anti-ER Ab-10 (Neomarkers) were incubated with the beads for 6 to 8 hours. Beads were washed 2x w/ PBS+BSA to remove unbound antibody, and resuspended in 0.5ml PBS+BSA. 280l of fixed chromatin (in sonication buffer) was mixed with 1.12 ml of Dilution Buffer (1% Triton X-100, 2mM EDTA, 150 mM NaCl, 20 mM Tris-HCl pH 8.1) and added to the beads, followed by overnight incubation with rocking at 4oC. Beads were collected by magnetic pull down and washed 6X with RIPA buffer (50mM HEPES pH 7.5, 1mM EDTA pH 8.0, 0.7% Sodium Deoxycholate, 1% NP40, 0.5M LiCl) using 20 minute and 5 minute alternating wash times, and twice with TE for 5 minutes. DNA was released and cross-links reversed by resuspending in Elution Buffer (100mM NaHCO3, 1% SDS) and incubating at 65oC overnight (vortexing every 5 minutes for the first 30 minutes). Samples of input chromatin were also resuspended in elution buffer and incubated at 65oC to reverse crosslinks. DNA was purified using the Qiaquick PCR kit (Qiagen) and quantified using the Quant-IT dsDNA HS kit (Invitrogen).
4ng of input and ChIP samples from two biological replicates were prepared for sequencing essentially using the standard Illumina protocol, with the following modifications: we used the End-It kit (Epicentre Biotech.) for end repair, Taq polymerase for A addition, and Easy A thermostable polymerase (Agilent Technologies) for the final limited amplification. Ligated or final amplified products of between ~220 and ~320 bp size were isolated by E-gel or 1.6% agarose gel electrophoresis followed by Qiagen Gel Extraction kit purification. For all other steps, products were purified by addition of 30 mg glycogen (invitrogen #1286677) and EtOH precipitation, followed by a 70% EtOH wash. The library was quantified using the Quant-IT dsDNA HS kit (Invitrogen) and approximately 10 ng submitted for sequencing. First replicate samples were not multiplexed. 2nd replicate samples were multiplexed using Illumina multiplex adapters and run together on one HiSeq lane (Aorta_IP=Barcode 2, Aorta INPUT=barcode 6, Liver IP=Barcode 4, Liver input= barcode 8). Raw data files were de-multiplexed by the sequencing facility.

Library strategy

ChIP-Seq

Library source

genomic

Library selection

ChIP

Instrument model

Illumina HiSeq 1000

Description

LiE_IP_eqnum_OnN_nwst_28bp_dedup.bed.gz
LiE_man2_n_newest3_eqnum_dup1_nomod148_p1e-4_treat_afterfiting_all.bdg.gz

Data processing

Reads were mapped to the mouse mm9 genome build using Bowtie version 0.12.5, with the following parameters --best --strata -n 1 (keeping alignments with at most 1 mismatch) -m 1 (keeping only reads that align to one location in the genome), and also trimmed to remove the first 8 bp (bowtie parameter -5 8) and trimmed on the 3' end to give a mapped read length of 28 bp (e.g. for 51 bp reads, -5 8 and -3 15 resulted in 28 bp read lengths). All mapped reads for each sample were concatenated together, sorted and deduplicated (keeping only one of any set of reads starting at the exact same base). Finally, an equal number of deduplicated reads from replicate 1 and replicate 2 were combined, to ensure equal weighting of each replicate (accomplished by randomly subsampling the longer file and concatenating it with the smaller file). The final counts of reads were 33.4 million, 70.1 million, 44.5 million and 73.9 million for the combined aorta ChIP, aorta input, liver ChIP and liver input samples, respectively.
ERa binding peaks that were significantly different from the input control were called using MACs version 1.4, a Poisson-based algorithm which evaluates the likelihood of ERa binding at a specific site over random chance, with the following parameters: --format=BED --keep-dup=1 --nomodel --bw=200 --shiftsize=100 --pvalue=1e-4 -B -S -c INPUT.bed -t ERaIP.bed --name output_prefix. We found that we got the best ratio of positive to negative(false) peaks when we did not use MACS to estimate fragment size, but instead used experimentally determined fragment sizes (using the MACS parameter --nomodel and supplying fragment size and 1/2 fragment size values as --bw and --shiftsize). Fragment sizes were 200 for the Aorta IP and 148 for the Liver IP.
Peaks were called as high probability ERalpha binding sites if they showed >=3-fold enrichment over background, p<=10-5 and empirical FDR of <=10%. With these thresholds we found 1754 high quality ERBSes in aorta, and 33273 in liver.
Genome_build: mm9
Supplementary_files_format_and_content: Two types of processed files are included: 1) .bed files containing an equal number of mapped, deduplicated reads from replicate 1 & 2 of each sample type (input data to MACS, whose inclusion is important because of the random subsampling we used to give equal read numbers for each replicate) and 2) .wig files of shifted read densities for input and IP data (output from MACS).
Supplementary_files_format_and_content: peak file containing peak calls (p. <1e-5, FDR<10%, Fold-enrichment > 3)

Submission date

Nov 13, 2013

Last update date

May 15, 2019

Contact name

Gavin R. Schnitzler

E-mail(s)

gavin.schnitzler@gmail.com

Phone

6086950428

Organization name

Tufts Medical Center