WashU Epigenome Browser

GSM500898_4hr_K4_rep1_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500899_4hr_K27_rep1_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500900_4hr_K36_rep1_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500901_4hr_polII_rep1_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500902_4hr_K4_rep2_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500903_4hr_K36_rep2_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500904_4hr_K27_rep2_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500905_4hr_polII_rep2_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500906_256_K4_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500907_256_K27_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500908_256_K36_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM500909_256_polII_MA2Cscore	Detail=Each array was normalized by MA2C. Each probe was assigned a MA2Cscore to reflect the normalized and window averaged log2 ratio of ChIP enrichment over control (see MA2Cscore.wig supplementary file).

GSM564427_neg_RZN001_score22_zv9_density	Detail=ABI pipeline BioScope v1.0.1, Data analysis: The SOLiD-generated RNA-Seq reads was in 50bp length and an initial filtering process was taken to remove any non-desirable contamination sequences, such as rRNA, tRNA, and repeats etc. A seed and extension mapping approach was developed to map the 50bp reads into reference genome zv7 and also into the respective zv7 refGene annotation separately. During mapping of the reads to the refGene annotation, a splice junction database fasta file is generated to which the reads are mapped to. This database contains known and putative junction sequences, created by taking 46bp from the joining ends of adjacent and non-adjacent exons for each gene and putting them together, simulating the genomic sequences of known and putative junctions respectively. The first 25bp of the 50bp read was used as the seed for alignment for each read. When an alignment cannot be found using these 25bp seed, the seed window is shifted and the next 25bp seed is taken from the 21st to the 45th base of the read. An extension step follows when the seed is able to map, and a score generated for each extended base to determine best alignment. A merging step is performed on the genome mapping and splice junction mapping to determine a set of alignments for each read, from which a unique alignment is found based on the score generated for these alignments., Mapping parameters:  Mapping was done using Applied Biosystems’ SOLiD BioScope alignment for whole transcriptome analysis pipeline.  Two mismatches were allowed in the 25bp color space seed sequence with extension alignment performed to find the full mapping location.  A score is computed for each mapping location and any location that scored <22 were filtered., Generation of gff and bedgraph files: GFF files were produced by parsing the output (BAM format) and extracted for unique alignments with score >22. The bedgraph files are then produced with the resulting gff file.

GSM564428_neg_RZN002_score22_zv9_density	Detail=ABI pipeline BioScope v1.0.1, Data analysis: The SOLiD-generated RNA-Seq reads was in 50bp length and an initial filtering process was taken to remove any non-desirable contamination sequences, such as rRNA, tRNA, and repeats etc. A seed and extension mapping approach was developed to map the 50bp reads into reference genome zv7 and also into the respective zv7 refGene annotation separately. During mapping of the reads to the refGene annotation, a splice junction database fasta file is generated to which the reads are mapped to. This database contains known and putative junction sequences, created by taking 46bp from the joining ends of adjacent and non-adjacent exons for each gene and putting them together, simulating the genomic sequences of known and putative junctions respectively. The first 25bp of the 50bp read was used as the seed for alignment for each read. When an alignment cannot be found using these 25bp seed, the seed window is shifted and the next 25bp seed is taken from the 21st to the 45th base of the read. An extension step follows when the seed is able to map, and a score generated for each extended base to determine best alignment. A merging step is performed on the genome mapping and splice junction mapping to determine a set of alignments for each read, from which a unique alignment is found based on the score generated for these alignments., Mapping parameters:  Mapping was done using Applied Biosystems’ SOLiD BioScope alignment for whole transcriptome analysis pipeline.  Two mismatches were allowed in the 25bp color space seed sequence with extension alignment performed to find the full mapping location.  A score is computed for each mapping location and any location that scored <22 were filtered., Generation of gff and bedgraph files: GFF files were produced by parsing the output (BAM format) and extracted for unique alignments with score >22. The bedgraph files are then produced with the resulting gff file.

GSM564429_neg_RZN003_score22_zv9_density	Detail=ABI pipeline BioScope v1.0.1, Data analysis: The SOLiD-generated RNA-Seq reads was in 50bp length and an initial filtering process was taken to remove any non-desirable contamination sequences, such as rRNA, tRNA, and repeats etc. A seed and extension mapping approach was developed to map the 50bp reads into reference genome zv7 and also into the respective zv7 refGene annotation separately. During mapping of the reads to the refGene annotation, a splice junction database fasta file is generated to which the reads are mapped to. This database contains known and putative junction sequences, created by taking 46bp from the joining ends of adjacent and non-adjacent exons for each gene and putting them together, simulating the genomic sequences of known and putative junctions respectively. The first 25bp of the 50bp read was used as the seed for alignment for each read. When an alignment cannot be found using these 25bp seed, the seed window is shifted and the next 25bp seed is taken from the 21st to the 45th base of the read. An extension step follows when the seed is able to map, and a score generated for each extended base to determine best alignment. A merging step is performed on the genome mapping and splice junction mapping to determine a set of alignments for each read, from which a unique alignment is found based on the score generated for these alignments., Mapping parameters:  Mapping was done using Applied Biosystems’ SOLiD BioScope alignment for whole transcriptome analysis pipeline.  Two mismatches were allowed in the 25bp color space seed sequence with extension alignment performed to find the full mapping location.  A score is computed for each mapping location and any location that scored <22 were filtered., Generation of gff and bedgraph files: GFF files were produced by parsing the output (BAM format) and extracted for unique alignments with score >22. The bedgraph files are then produced with the resulting gff file.

GSM564430_neg_RZN004_score22_zv9_density	Detail=ABI pipeline BioScope v1.0.1, Data analysis: The SOLiD-generated RNA-Seq reads was in 50bp length and an initial filtering process was taken to remove any non-desirable contamination sequences, such as rRNA, tRNA, and repeats etc. A seed and extension mapping approach was developed to map the 50bp reads into reference genome zv7 and also into the respective zv7 refGene annotation separately. During mapping of the reads to the refGene annotation, a splice junction database fasta file is generated to which the reads are mapped to. This database contains known and putative junction sequences, created by taking 46bp from the joining ends of adjacent and non-adjacent exons for each gene and putting them together, simulating the genomic sequences of known and putative junctions respectively. The first 25bp of the 50bp read was used as the seed for alignment for each read. When an alignment cannot be found using these 25bp seed, the seed window is shifted and the next 25bp seed is taken from the 21st to the 45th base of the read. An extension step follows when the seed is able to map, and a score generated for each extended base to determine best alignment. A merging step is performed on the genome mapping and splice junction mapping to determine a set of alignments for each read, from which a unique alignment is found based on the score generated for these alignments., Mapping parameters:  Mapping was done using Applied Biosystems’ SOLiD BioScope alignment for whole transcriptome analysis pipeline.  Two mismatches were allowed in the 25bp color space seed sequence with extension alignment performed to find the full mapping location.  A score is computed for each mapping location and any location that scored <22 were filtered., Generation of gff and bedgraph files: GFF files were produced by parsing the output (BAM format) and extracted for unique alignments with score >22. The bedgraph files are then produced with the resulting gff file.

GSM564431_neg_RZN005_score22_zv9_density	Detail=ABI pipeline BioScope v1.0.1, Data analysis: The SOLiD-generated RNA-Seq reads was in 50bp length and an initial filtering process was taken to remove any non-desirable contamination sequences, such as rRNA, tRNA, and repeats etc. A seed and extension mapping approach was developed to map the 50bp reads into reference genome zv7 and also into the respective zv7 refGene annotation separately. During mapping of the reads to the refGene annotation, a splice junction database fasta file is generated to which the reads are mapped to. This database contains known and putative junction sequences, created by taking 46bp from the joining ends of adjacent and non-adjacent exons for each gene and putting them together, simulating the genomic sequences of known and putative junctions respectively. The first 25bp of the 50bp read was used as the seed for alignment for each read. When an alignment cannot be found using these 25bp seed, the seed window is shifted and the next 25bp seed is taken from the 21st to the 45th base of the read. An extension step follows when the seed is able to map, and a score generated for each extended base to determine best alignment. A merging step is performed on the genome mapping and splice junction mapping to determine a set of alignments for each read, from which a unique alignment is found based on the score generated for these alignments., Mapping parameters:  Mapping was done using Applied Biosystems’ SOLiD BioScope alignment for whole transcriptome analysis pipeline.  Two mismatches were allowed in the 25bp color space seed sequence with extension alignment performed to find the full mapping location.  A score is computed for each mapping location and any location that scored <22 were filtered., Generation of gff and bedgraph files: GFF files were produced by parsing the output (BAM format) and extracted for unique alignments with score >22. The bedgraph files are then produced with the resulting gff file.

GSM564432_neg_RZN006_score22_zv9_density	Detail=ABI pipeline BioScope v1.0.1, Data analysis: The SOLiD-generated RNA-Seq reads was in 50bp length and an initial filtering process was taken to remove any non-desirable contamination sequences, such as rRNA, tRNA, and repeats etc. A seed and extension mapping approach was developed to map the 50bp reads into reference genome zv7 and also into the respective zv7 refGene annotation separately. During mapping of the reads to the refGene annotation, a splice junction database fasta file is generated to which the reads are mapped to. This database contains known and putative junction sequences, created by taking 46bp from the joining ends of adjacent and non-adjacent exons for each gene and putting them together, simulating the genomic sequences of known and putative junctions respectively. The first 25bp of the 50bp read was used as the seed for alignment for each read. When an alignment cannot be found using these 25bp seed, the seed window is shifted and the next 25bp seed is taken from the 21st to the 45th base of the read. An extension step follows when the seed is able to map, and a score generated for each extended base to determine best alignment. A merging step is performed on the genome mapping and splice junction mapping to determine a set of alignments for each read, from which a unique alignment is found based on the score generated for these alignments., Mapping parameters:  Mapping was done using Applied Biosystems’ SOLiD BioScope alignment for whole transcriptome analysis pipeline.  Two mismatches were allowed in the 25bp color space seed sequence with extension alignment performed to find the full mapping location.  A score is computed for each mapping location and any location that scored <22 were filtered., Generation of gff and bedgraph files: GFF files were produced by parsing the output (BAM format) and extracted for unique alignments with score >22. The bedgraph files are then produced with the resulting gff file.

RNA_2-4cell	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

RNA_1Kcell	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

RNA_dome	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

RNA_shield	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

RNA_bud	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

RNA_28hpf	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

RNA_2dpf	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

RNA_5dpf	Detail=Th summary result files of he developmental transcriptome of all samples are  available as supplementary information with the paper.

H3K4me3_shield	Detail=Reads were mapped to the Zv9 genome assembly using bowtie

H3K27me3_shield	Detail=Reads were mapped to the Zv9 genome assembly using bowtie

ChIP_WCE_shield	Detail=Reads were mapped to the Zv9 genome assembly using bowtie

GSM675174_MBT_H3K27me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675175_MBT_H3K27me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675176_MBT_H3K36me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675177_MBT_H3K36me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675178_MBT_H3K4me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675179_MBT_H3K4me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675180_MBT_H3K9me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675181_MBT_H3K9me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675182_MBT_polii_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675183_MBT_polii_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675184_postMBT_H3K27me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675185_postMBT_H3K27me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675186_postMBT_H3K36me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675187_postMBT_H3K36me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675188_postMBT_H3K4me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675189_postMBT_H3K4me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675190_postMBT_H3K9me3_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675191_postMBT_H3K9me3_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675192_postMBT_polii_1	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675193_postMBT_polii_2	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675194_preMBT_H3K27me3	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675195_preMBT_H3K36me3	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675196_preMBT_H3K4me3	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675197_preMBT_H3K9me3	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM675198_preMBT_polii	Detail=log2 (ChIP/Input) with biweight mean of values subtracted

GSM813749_H3K4me3_IP_24hpf	Detail=Reads were aligned to the zebrafish danRer7 genome assembly using Bowtie, allowing for up to one mismatch and up to 4 genomic matches. Peak detection was performed with the Model-based Analysis of ChIP-Seq (MACS 1.4) algorithm (http://liulab.dfci.harvard.edu/MACS/), using the Input libraries to assess FDR

GSM813750_H3K4me3_IP_72hpf	Detail=Reads were aligned to the zebrafish danRer7 genome assembly using Bowtie, allowing for up to one mismatch and up to 4 genomic matches. Peak detection was performed with the Model-based Analysis of ChIP-Seq (MACS 1.4) algorithm (http://liulab.dfci.harvard.edu/MACS/), using the Input libraries to assess FDR

GSM813751_H3K4me3_IP_Adult	Detail=Reads were aligned to the zebrafish danRer7 genome assembly using Bowtie, allowing for up to one mismatch and up to 4 genomic matches. Peak detection was performed with the Model-based Analysis of ChIP-Seq (MACS 1.4) algorithm (http://liulab.dfci.harvard.edu/MACS/), using the Input libraries to assess FDR

GSM813752_H3K36me3_IP_24hpf	Detail=Reads were aligned to the zebrafish danRer7 genome assembly using Bowtie, allowing for up to one mismatch and up to 4 genomic matches. Peak detection was performed with the Model-based Analysis of ChIP-Seq (MACS 1.4) algorithm (http://liulab.dfci.harvard.edu/MACS/), using the Input libraries to assess FDR

GSM813753_H3K36me3_IP_72hpf	Detail=Reads were aligned to the zebrafish danRer7 genome assembly using Bowtie, allowing for up to one mismatch and up to 4 genomic matches. Peak detection was performed with the Model-based Analysis of ChIP-Seq (MACS 1.4) algorithm (http://liulab.dfci.harvard.edu/MACS/), using the Input libraries to assess FDR

GSM813754_H3K36me3_IP_Adult	Detail=Reads were aligned to the zebrafish danRer7 genome assembly using Bowtie, allowing for up to one mismatch and up to 4 genomic matches. Peak detection was performed with the Model-based Analysis of ChIP-Seq (MACS 1.4) algorithm (http://liulab.dfci.harvard.edu/MACS/), using the Input libraries to assess FDR

GSM853234_Nanog_3.5hpf	Detail=Nanog_ZV8Embryo3.5.ylf, genome build: zv8, Nanog_ZV8Embryo3.5.WIG.gz, genome build: zv8, Counts: Sequence reads were obtained and mapped to the zebrafish Zv8 genome using the Illumina Genome Analyzer Pipeline. All reads mapping with two or fewer mismatches were retained and read starts were summed in sliding windows of 400 bp to create summary windows. Counts included.

GSM853235_Nanog_4.5hpf	Detail=Nanog_ZV8Embryo4.5.ylf, genome build: zv8, Nanog_ZV8Embryo4.5.WIG.gz, genome build: zv8, Counts: Sequence reads were obtained and mapped to the zebrafish Zv8 genome using the Illumina Genome Analyzer Pipeline. All reads mapping with two or fewer mismatches were retained and read starts were summed in sliding windows of 400 bp to create summary windows. Counts included.

GSM853236_Mxtx_4.5hpf	Detail=Mxtx_ZV8Embryo4.5.ylf, genome build: zv8, Mxtx_ZV8Embryo4.5.WIG.gz, genome build: zv8, Counts: Sequence reads were obtained and mapped to the zebrafish Zv8 genome using the Illumina Genome Analyzer Pipeline. All reads mapping with two or fewer mismatches were retained and read starts were summed in sliding windows of 400 bp to create summary windows. Counts included.

H3K4me3_dome.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K4me3_80epi.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K4me3_24hpf.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K4me3_48hpf.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K4me1_dome.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K4me1_80epi.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K4me1_24hpf.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K4me1_48hpf.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K27ac_dome.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K27ac_80epi.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K27ac_24hpf.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

H3K27ac_48hpf.norm	Detail=Reads were mapped to the D. rerio genome (ENSEMBL version Zv9) with the ELAND software (from Illumina CASAVA 1.8.2) using default settings. Reads with 0 or 1 mismatch were kept.

GSM822655_preMBT_MeDIP_1	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM822656_preMBT_MeDIP_2	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM822657_MBT_MeDIP_1	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM822658_MBT_MeDIP_2	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM822659_postMBT_MeDIP_1	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM822660_postMBT_MeDIP_2	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM822661_ZF4_MeDIP_1	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM822662_ZF4_MeDIP_2	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM913999_sperm_MeDIP_1	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM914000_sperm_MeDIP_2	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM914001_emb24h_MeDIP_1	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM914002_emb24h_MeDIP_2	Detail=log2 (MeDIP/Input) with biweight mean of values subtracted

GSM1133391_chrall.sperm.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133392_chrall.egg.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133393_chrall.16-cell.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133394_chrall.32-cell.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133395_chrall.64-cell.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133396_chrall.128-cell.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133397_chrall.1k-cell.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133398_chrall.Germring.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site

GSM1133399_chrall.testis.bedGraph	Detail=Illumina Casava1.7 software used for basecalling., Sequenced reads were trimmed for adaptor sequence, and masked for low-complexity or low-quality, Zv9 reference genome was downloaded from UCSC. The 48,502 bp Lambda genome was also included in the reference sequence as an extra chromosome for calculating bisulfite conversion rate. Filtered paired-end methylC-seq reads were mapped against the reference by Bismark_v0.6.4 (Krueger and Andrews, 2011) with stringent parameters:  -n 2 -l 60  -e 100 -X 600. A custom script was used to examine whether pair-end reads were overlapped and the overlapped part is trimmed from one end to prevent counting twice from the same observation., For CpG i, define mi as the number of reads showing methylation over position i (both strands). Define ui as the number of reads showing lack of methylation over CpG i. The methylation level is estimated as mi/( mi + ui), which is an estimate of the probability that CpG i is methylated in a molecule sampled randomly from the cell population. Because CpG methylation is symmetric, mi and ui include observations associated with the cytosines on both strands for the i-thCpG., for TAB-seq , 5-hmC mapping was performed as described (Yu et al., 2012). Paired-reads were mapped uniquely to the reference genome (Zv9, USCS) plus Lambda and pUC19 DNA sequence as extra chromosome by Bismark. All the CpG sites disrupted by SNPs were filtered from the analysis. 5hmC was called with a binomial distribution follows the method previously reported (Yu et al., 2012). A binomial distribution (Lister et al., 2009)was applied to model this probabilistic event with N as the depth of sequencing at the cytosine and p as the 5mC non-conversion rate. Efficient conversion of unmodified cytosine to uracil and efficient conversion of 5mC to 5caU/U were calculated by spiked M. SsI treated lambda DNA., Genome_build: Zv9/danRer7, Supplementary_files_format_and_content: wig files contain  methylaion level of each CpG site