ChIP-BIT2 was implemented using C/C++. The pipeline of ChIP-BIT2 was shown in Fig. 2 (Additional file 1: Fig. S2). Given a pair of sample and input ChIP-seq profiles in SAM format, ChIP-BIT2 firstly extracted the genomic coordinates of individual reads from sample and input ChIP-seq profiles, respectively (Additional file 1: Fig. S3). And then it detected peaks at promoters, enhancers (if annotation files were provided), or across the whole genome.

ChIP-BIT2 pipeline. ChIP-BIT2 respectively extracted read location information from sample and input ChIIP-seq SAM format profiles. Depending on the running mode, it can detect peaks from the whole genome or from annotated regulatory regions like promoters or enhancers. To enable  peak detection of different sizes, ChIP-BIT2 partitioned genomic segments into smaller windows and calculated read intensity in each window for distribution parameter learning and binding occurrence probability estimation. Windows with the posterior probability over 0.9 were output in BED format as final peaks

Promoters refer to the proximal regulatory regions centered around gene TSS. A TSS annotation file is required to enable the ‘-promoter’ running mode of ChIP-BIT2. Users can set the preferred promoter size using the ‘-s’ option. Under this mode, ChIP-BIT2 jointly modeled read intensities in the sample and input ChIP-seq profiles using the Gaussian mixture model [Eq. (2)]. In the meanwhile, it modeled the relative distance of each window to the nearest TSS using the Exponential-Uniform mixture model [Eq. (3)]. A demo of using ChIP-BIT2 for detecting peaks at promoters was provided in Additional file 1: Fig. S4.

Enhancers referred to distal regulatory regions interacting with promoters/TSSs in the 3D genome [18]. The linear distance of an enhancer to its target promoter/gene can be up to 1 Mbps. Some proteins like EP300, H3K27ac, and H3K4me1 specifically bind to enhancers frequently and have higher ChIP-seq signal enrichment at enhancers than at promoters or other genomic locations [1012]. To effectively detect ChIP-seq peaks for such proteins, an enhancer annotation file is required to enable the ‘-enhancer’ running mode of ChIP-BIT2. Under this mode, ChIP-BIT2 modeled read intensities in the sample and input ChIP-seq profiles jointly using Eq. (2). Different from the promoter mode, ChIP-BIT2 modeled the relative distance of each window to the enhancer center using a Uniform distribution [Eq. (4)]. A demo of using ChIP-BIT2 for peak detection at enhancers was provided in Additional file 1: Fig. S5.

Promoters and enhancers are two categories of well-understood regulatory regions. There exist many other types of genomic regions also bound by DNA proteins. For example, cohesion proteins CTCF and RAD21 usually bind at the boundaries of topological associated domains and play a key role in the 3D chromatin structure [19]. Transcription initiation protein POLA2 binds to all active regulatory regions in the whole genome. For such proteins, it is important to call their ChIP-seq peaks from the whole genome, using the ‘-WG’ mode of ChIP-BIT2. As no annotated regulatory regions were needed, ChIP-BIT2 modeled read intensities from sample and input ChIP-seq profiles and predicted peaks at genome-wide locations.

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.