Cov2Coinfect

Here is the code and data for Hang-Yu Zhou, et al. "Genomic evidence for divergent co-infections of co-circulating SARS-CoV-2 lineages"

---

Overview

Hypergeometric Distribution

where

• N is the population size (the total number of nonsynonymous mutations that occur in 2.5 million SARS-CoV-2 consensus genomes),
• K is the number of success states in the population (the number of feature variations of one lineage),
• n is the number of draws (the number of remaining undefined mutations of sample),
• k is the number of observed successes (the number of remaining undefined mutations that occur both in sample and lineage feature variations)

For each lineage, hypergeometric test computes the probability (p-value) of observed successes (mutations that occurred both in sample and lineage feature variations) under “null hypothesis”: the hypothesis that there is nothing special about the lineage. If the p-value is sufficiently low, we can reject the null hypothesis as too impossible and conclude that the sample are highly correlated with the tested lineage.

Statistical Significance

• Lineages with lower p-value are more likely assigned to the sample.

Mutation Frequency Uniformity

• Frequencies of mutations that occur both in sample and one lineage feature variations should have a standard deviation less than 20.

Mutation Concentration

• Number of mutations that occur both in sample and one lineage feature variations should be greater than 6.
• Number of mutations that occur both in sample and one lineage feature variations divided by number of the lineage feature variations should be greater than 0.3.

Usage

1. Detect variant in SRA file using wuaipinglab/sra2variant.

2. Run get_lineagesFV_and_mutationNum.py to get lineages feature variations and number of global SARS-CoV-2 nonsynonymous mutations.

3. Run get_candidate_lineages.py to get candidate with-in host lineages.

4. Run identify_co-infection_lineages.py to get defined with-in host lineages and identify co-infection cases.

Citation

Genomic evidence for divergent co-infections of co-circulating SARS-CoV-2 lineages by Hang-Yu Zhou, et al. Computational and Structural Biotechnology Journal