Best Practices for chi-seq analysis

Question

[luckysardar171]

Can anyone suggest me best practices for chip-seq analysis complete pipeline, I am new to chip-seq analysis.

Thank you

Answer 1

[predeus]

There are two big classes of problems - narrow peak and broad peak Chip-seq.

For narrow peaks, classic analysis with alignment by any good genome-wide aligner (e.g. bowtie2) and peak calling by MACS2 should work well. Downstream analysis tends to be focused on motifs, which can mostly (in first approximation) be covered by MEME suite.

For broad peaks (e.g. many histone modifications), it's harder to pinpoint the best "go-to" peak caller. SICER worked quite nicely, and has been re-implemented several times by different groups. Downstream analysis tends to be focused on HMM-based segmentation, e.g. with ChromHMM.

Analysis also changes quite a bit if you have biological replicates, or need to do comparison between conditions. There's lot's of experimental approaches, e.g. SPAN.

[luckysardar171]

Any pipeline for chip_seq?

[predeus]

Take a look at what ENCODE did for a more complete pipelines.