Shulou(Shulou.com)06/01 Report--

How to use peak calling software MACS, aiming at this problem, this article introduces the corresponding analysis and answer in detail, hoping to help more partners who want to solve this problem to find a more simple and feasible method.

MACS, whose full name is Model-based Analysis of ChIP-Seq, is one of the most widely used peak calling software. its basic principles are briefly introduced as follows

1. Bias distance between Predictive sequencing reads and peak

In chip_seq and other data, scientists found that on both sides of the real binding site, the sequencing depth distribution of positive and negative strands is shown in the following figure, and the center of the corresponding peak is offset from the center of peak.

MACS first uses a model to evaluate the offset distance between the real peak center and the sequencing peak. Given the parameters bandwidth and mfold, a sliding window with a size of 2 bandwidth is used to compare the difference between the distribution of the real sequencing depth in the window and the random sequencing. If the difference multiple exceeds the threshold mfold, the window is considered to be a peak region. After the initial peak region is identified, 1000 highly reliable peak regions are randomly selected, and the sequencing depth distributions of positive and negative strands are calculated respectively, as shown below

The distance between the peaks of positive and negative chains is identified in this way, which is defined as d. In the case of subsequent peak calling, the distance of dmax 2 is offset to the 3 'end based on the initial calculation results.

2. Peak detection

It is generally considered that the distribution of sequencing depth on the genome is a Poisson distribution. According to the value of genome sequencing depth, the parameters of the whole Poisson distribution are estimated.

After the initial peak region is offset by the distance of dstroke 2, the sliding window which has been 2 days old is scanned again to find the enriched region. If the value of the sequencing distribution of a region is greater than that of the overall Poisson distribution, the region is considered to be enriched. For the enrichment regions of overlap, the enrichment regions were merged, and then starting from the central region of peak, the overall shift of d bases to the 3 'end, in which the point with the highest sequencing depth was taken as the peak summit of peak.

When comparing the differences between input and IP samples, MACS is also calculated based on Poisson distribution, but considering the influence of sequencing errors, repetitive regions, CNV and other factors, MACS adopts a dynamic Poisson distribution model, that is, it does not assume that all regions of the genome conform to the same Poisson distribution, but that the parameters of Poisson distribution can be different in different regions. For candidate peak regions, the parameters of Poisson distribution are estimated by the following model

1k represents the estimated value of the data in the 1k window centered on the peak center, and the following 5k and 10k are similar. Then the p value of each peak region is calculated by the parameters of the local Poisson distribution. If it is less than the threshold of the specified pvalue, the peak region is output, and the ratio of the number of IP samples in the region to the local Poisson distribution parameters is taken as the fold enrichment of the peak.

This is the answer to the question about how to use the peak calling software MACS. I hope the above content can be of some help to you. If you still have a lot of doubts to solve, you can follow the industry information channel to learn more about it.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.