Bacterial ChIPseq using 454?

[whitestein]

We are going to do a ChIPseq experiment in the next few months to find the targets of a transcription factor from our bacterial species of interest(~4000 genes).
I know that the Illumina platform is better for this type of experiment, but we have a GS Junior machine in-house and use of it could save us $$?

I figure with the reduced complexity of our samples that the GS Junior may be more than adequate for what we want to do. Or would we just be wasting our time?

Any advice is welcome.

[mps4208]

Many of my clients love the GS Junior, but it does have it's niche among sequencers. As long as you are aware of those at the beginning, you'll be happier with your outcome.

A few things to assess before using the GS Junior for this.
1) Are your fragments long enough? GS Junior requires longer fragments of dsDNA than other platforms. You will be needing your sample to be about 300-900bp. The Junior's optimum final read will be 400-550bp. Let your sequencing contact know if the sample is less than 1.5kb, so they don't have to do additional fragmentation.
2) Do your samples contain a lot of homopolymers? The GS Junior is a pyrosequencer and the technology doesn't do well with 5-7 of the same nucleotide.
3) GS Junior does require 500ng of double-stranded DNA, without any particulate matter, and a 260/280 ratio of 1.8.
4) GS Junior produces less reads than other platforms, on average ours only does 150-250K of reads per run.

If your sample meets these guidelines, then it should be a nice, money saving, fast option for you.

[whitestein]

Thanks for the info mps - super helpful. We are doing the fragmentation so we can aim for that size. The number of reads should be fine and I'm not worried about homopolymers because we have a genome sequence to compare to.

The 500 ng amount might be more difficult to achieve but I suppose we can scale up the size of our culture in a pilot experiment and see what we get.
Why does the GS junior require so much input DNA? Is that for optimum formulation with the beads? Could we get away with less?

[mps4208]

500ng used to be a small amount, but the other companies have managed to create more sensitive kits-Roche seems to be behind in that field.

500ng is because they assume that 1/2 is lost during fragmentation. Since you are doing your own fragmentation, you should be able to get away with less, but talk to whoever is running the instrument. We have started doing ds cDNA samples on ours and can accept 100-300ng since we don't have to fragment, but it just barely seems to pass QC at that amount.