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

What is the application of big data negative binomial distribution in difference analysis? in view of this problem, this article introduces the corresponding analysis and solution in detail, hoping to help more partners who want to solve this problem to find a more simple and feasible method.

Why should it be based on negative binomial distribution?

From a statistical point of view, difference analysis will definitely need hypothesis testing, usually for the data whose distribution is known, the false positive rate of parameter test results will be lower. What is the distribution of raw count values in transcriptome data?

The count value is essentially the number of reads, is a non-zero integer, and is discrete, and its distribution must also be discrete. For transcriptome data, the commonly used distributions in academic circles include Poisson distribution and negative binomial distribution.

In the early stage of data analysis, some scholars did use Poisson distribution for difference analysis, but up to now, almost all of them are based on negative binomial distribution. What factors lead to this phenomenon? To explain this, we must mention the concept overdispersion.

Dispersion refers to the degree of discreteness, the study of the degree of discretization of a data distribution, we often use variance as an indicator. For Poisson distribution, the mean and variance are equal, but our data do not conform to this law. By calculating the mean and variance of all genes, you can draw the following picture

Abscissa is the mean of genes in all samples, ordinate is the variance of genes in all samples, and the slope of the straight line is 1, which represents the distribution of mean and variance of Poisson distribution. It can be seen that the distribution of the real data deviates from the Poisson distribution, and the variance is obviously larger than the mean.

The corresponding code for the above picture is as follows

Mean

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.