[JohnK]

The gene shown is one that we have an interest in but isn't one that is worked on in the lab specifically. Unlikely that anyone would have a prep of this floating around but I can check - I'm not the wet lab person.

If there is no obvious explanation, I guess the picture is intriguing enough to warrant some lab investigations to figure out if it's an artefact or not. I thought there might be some previous experience of such an occurrence.

Did you de-dup?

[pogaora]

Did you de-dup?

Yes indeed.

[JohnK]

Yes indeed.

Possibility:

Bias towards GC-rich content in fragment selection both in library prep and amplification before and during sequencing.

(Park, October 2009)

[mudshark]

as there are no such peaks in the input I assume that it is not a systematic bias such as an over representation of GC rich fragments

[Chipper]

We haven't done any exome capture experiments, but there were RNAseq samples on the same flowcell. The picture shown is not typical of most genes though - even some which we know are highly expressed in these cells.

If RNA-seq was strand-specific it may explain why you did not get any peak calls (if your peak caller requires peaks on both strands). You could perhaps calculate RPKMs and compare to the other samples, or compare SNP calls in exons to figure out where the reads come from...

Bias may be different in ChIP and input since you start with less material in the ChIP library, but GC-bias will not result in such enrichments over exons.

[JohnK]

I talked to a colleague who sees these exon-peak events in some of his data depending on what protein or transcription factor he's after. They could be perfectly valid.

[ewilbanks]

Hi folks,

Is your target protein expected to bind to a single location (like a transcription factor)? or over a more distributed region?

If the former is true, I would suggest visualizing your read pile-up using the Broad's IGV which lets you look nicely at the strand-bias (Fig 1 here for an explanation of what I mean!). I've found this to be a great way visually inspecting the data for true binding sites vs. noise-peaks.

Cheers,
Lizzy

[pogaora]

I talked to a colleague who sees these exon-peak events in some of his data depending on what protein or transcription factor he's after. They could be perfectly valid.

Hello again all. Thanks to everyone for their interest in this question.

Does the above mean that your colleague validated in the wet lab?

The peaks we're talking about are from targetting a transcription factor and we see reads on both strands in all of the peaks that were in the pic posted earlier.

[yxi]

Hi Pogaora,

We observed the same exon enriched peak patterns in our ChIP-seq datasets. We also tried another replicate but it didn't happen again. Can you reproduce such peak patterns? I'm interested in comparing your data with ours (we can provide our data too if you are interested). I've sent you a message about the details.

Best regards,

yxi

[steven]

Not sure it helps, but this previous thread looks related.

[hon]

It may also due to RNAase treatment is not enough, so mRNA was carried over to library construction. Make sure RNAase treatment is done long enough, trace of RNA will be magnified in library.

[mudshark]

It may also due to RNAase treatment is not enough, so mRNA was carried over to library construction. Make sure RNAase treatment is done long enough, trace of RNA will be magnified in library.

how should that work technically, mRNA getting fixed into the library? you have any experimental evidence?

[hon]

I didn't have exact experimental evidence but that phenomenon also occurred to me once that many reads pile up on exons of some genes. My reason is that only mRNA mapped to exons, it was either due to mRNA contamination in the lab or ChIP sample was not completely RNAase treated. Since then, I paid careful attention to ChIP-seq library contamination, it didn't happen again.