Percent of ChIP-seq reads mapped

[kmcarr]

We just ran a ChIP-seq sample on our GAII, just one lane. The sample was from a prokaryote and all indications are that the library prep and sequencing went well. We got ~16 million clusters, 8.6 million passed the default filtering. When we aligned the reads to the reference genome only ~50% of the filter passed reads aligned. We tried both Eland and MAQ, using default settings for both. We are just starting to work with ChIP so we don't have much in house experience to compare to. I guess my first questions are, what percentage of ChIP-seq reads are you typically able to align to your reference? Do you make any special adjustments to alignment parameters when working with ChIP data?

Thanks in advance.

[dcfargo]

We get 40-60% of the total CHIP reads to align to unique genomic sequence (from flys).

[huguesparri]

Same for us.
We have 50-60% reads passing the filtering and 60-70% of them being aligned uniquely on the ref. genome.
In the end, 40 to 50% of the initial number of tag (raw data) gets aligned uniquely to the ref. genome.
And we've tried many organisms: mouse, human, fly,...
For the moment, we didn't notice a correlation between the number of uniquely aligned tags and the organism we're working on.