Recently, one of my colleagues has found that histone modifications cover up to 15% of the human genome, but for all different cell types tested they cover only up to 8% in mice. She wants to know if this is a result of the different experiment setups or if this is true. Any ideas?
Your best start point for data will be the ENCODE and mouse ENCODE projects (http://genome.ucsc.edu/ENCODE/dataMatrix/encodeChipMatrixHuman.html and http://genome.ucsc.edu/ENCODE/dataMatrix/encodeChipMatrixMouse.html). These projects characterized many (both different and redundant) histone modifications in a large collection of cell types (mainly in-vitro cell lines).
Though I can not exclude small differences between organisms, my first thought is that this difference is largely experimental. The question here is, what is coverage? In all experiments, you will have to set a threshold between positive and negative. Does this mean that all nucleosomes in the negative fraction do not carry the mark, and do all nucleosomes in the positive fraction?
Most likely it's not that black and white. Besdies that, there are many variables: which technique is used (ChIP-seq has a different signal to noise ratio as ChIP-on-chip), which antibodies (signal-to-noise can vary tremendously), were the samples processed identically (eg. with or without crosslinking), how homogeneous were the cell populations, were the cells stressed, or are these cells dividing fast or senescent (all associated with specific patterns of histone modifications).
Unless you control your experiments very well (which I think is beyond our current capabiliities), it will be very difficult to assign such a strict number. Still, very drastic differences between cell types can be observed at different positions in the genome, and these may directly relate to biological functions.
I am not sure what you mean by this? Histone modifications cover much more than 15% of the human genome. Indeed, most every nucleosome is likely to have some modifications. Please remember, that when one does a whole genome scan for modifications, it is generally only the peaks above background that register and are studied. The rest of the genome, i.e. the non-peaks, still have modifications of one type or another. If they don't have H3K4me, they likely have H3K27me, or H4K20 or something else. To suggest that there are no modifications on 85% of chromatin I think is incorrect.
You can try EpiExplorer (Genome Biology 2012, http://epiexplorer.mpi-inf.mpg.de). This tool allows to make query on both human and mouse epigenomes.
@Daan Noordermeer
thanks for your information.
"Though I can not exclude small differences between organisms, my first thought is that this difference is largely experimental."
Would you please give an example paper dealing with this difference (especially the genome coverage) of specific histone modification between organisms?
"The question here is, what is coverage?"
By coverage I mean the the total genome coverage
(in bp overlapping with a histone modification divided by total bp)
And I agreed with your "not that black and white" theory and it does make sense.
@Gregory Dressler
sorry for the confusion. By "Histone modifications cover much more than 15% of the human genome", I mean a specific histone modification (say H3K4me3) covers 15% of the human genome (the total bases of peak divided by the total bases of the genome).
And Do you know any papers about the coverage difference of a specific histone modification in mouse and human?
- Badges
- Science topic
- Similar topics
- Biological Science
- Genetics