Hi all,
I'm working with an Estrogen Receptor Chip-seq data set and I'm using the SPP phantom peak quality tools to calculate cross correlation and the plots produced look very bizarre compared to the ENCODE standards (from this paper).
ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia
(DOI: 10.1101/gr.136184.111)
This study is following a study using tissue derived from mouse uterus where they treated with E2 and performed ChIP for ER alpha and Pol2.
(DOI: 10.1210/me.2011-1311)
I've downloaded their datasets from SRA and performed the same cross correlation analysis and - to put it shortly - their data looks normal, whereas our data looks kind of crazy. The mapping was performed using the same programs and my reads were quality filtered and the mapping results look the same. The SRA data was mapping using the --sra-acc option in hisat2, so no filtering was necessary.
I've linked our cross correlation plots for reference.
Plot 1:
Plot 2:
Published Uterine ChIPseq data from SRA:
ER ChIP no treatment
I was hoping someone might be able to help me interpret what I'm looking at... and whether or not the data set is still usable for analysis. In particular, what would cause so many jagged peaks across the entire plot??
Thanks in advance,
Paul
I'm working with an Estrogen Receptor Chip-seq data set and I'm using the SPP phantom peak quality tools to calculate cross correlation and the plots produced look very bizarre compared to the ENCODE standards (from this paper).
ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia
(DOI: 10.1101/gr.136184.111)
This study is following a study using tissue derived from mouse uterus where they treated with E2 and performed ChIP for ER alpha and Pol2.
(DOI: 10.1210/me.2011-1311)
I've downloaded their datasets from SRA and performed the same cross correlation analysis and - to put it shortly - their data looks normal, whereas our data looks kind of crazy. The mapping was performed using the same programs and my reads were quality filtered and the mapping results look the same. The SRA data was mapping using the --sra-acc option in hisat2, so no filtering was necessary.
I've linked our cross correlation plots for reference.
Plot 1:
Plot 2:
Published Uterine ChIPseq data from SRA:
ER ChIP no treatment
I was hoping someone might be able to help me interpret what I'm looking at... and whether or not the data set is still usable for analysis. In particular, what would cause so many jagged peaks across the entire plot??
Thanks in advance,
Paul