anvi-get-pn-ps-ratio

Calculate the rates of non-synonymous and synonymous polymorphism for genes across environmetns using the output of anvi-gen-variability-profile..

🔙 To the main page of anvi’o programs and artifacts.

Authors

Can consume

contigs-db variability-profile-txt

Can provide

pn-ps-data

Usage

This program calculates the pN/pS ratio for each gene in a contigs-db and outputs it as a pn-ps-data artifact.

What is the pN/pS ratio?

The pN/pS ratio (first described in Schloissnig et al. 2012) is the ratio of 2 rates: the rates of non-synonymous (pN) and synonymous (pS) polymorphism. It is analogous to dN/dS, which is the ratio of rates between non-synonymous (dN) and synonymous substitutions between two strains. We calculate pN/pS from allele frequency obtained through SCVs and SAAVs. See the study by Kiefl et al. 2023 for additional information, and this reproducible workflow associated with that study to see use cases.

How do I use this program?

First, you will need to run anvi-gen-variability-profile using the flag --engine CDN to get a variability-profile-txt for SCVs (single codon variants), which we’ll name SCVs.txt in this example.

Then you can run this program like so:

A pN/pS value is calculated for each gene x sample combo. This will result in a directory called output_dir that contains several tables that describe each of your genes. See pn-ps-data for more information.

Interpreting the output

The output directory will contain five tab-delimited files. By default these are in long format (one row per gene-sample combination), but if you use the --pivot flag, they will be in matrix format (genes as rows, samples as columns).

Here is what each file contains:

• pNpS.txt: The pN/pS ratio for each gene in each sample. This is the primary output of the program. The general right-hand rule for a broad interpretation is that the values greater than 1 suggest diversifying selection, values less than 1 suggest purifying selection, and values around 1 suggest neutral evolution. Some entries may be empty, which means one of two things: (1) the gene had fewer SCVs than the --minimum-num-variants threshold set by default (default: 4), so the ratio was not considered reliable enough to report, or (2) both pN and pS were zero, making the ratio undefined. Values of inf indicate that pN was greater than zero but pS was zero (i.e., all observed polymorphisms were non-synonymous).

• pN.txt: The rate of non-synonymous polymorphism per non-synonymous site for each gene in each sample. This is the sum of non-synonymous fractions across all SCVs in a gene, normalized by the number of non-synonymous sites in that gene.

• pS.txt: The same as above but opposite as this one is the rate of synonymous polymorphism per synonymous site for each gene in each sample. This is the sum of synonymous fractions across all SCVs in a gene, normalized by the number of synonymous sites in that gene.

• num_SCVs.txt: The number of single codon variants observed for each gene in each sample. This is useful for understanding the statistical support behind each pN/pS estimate. When the --pivot flag is used, genes with no SCVs in a given sample will show 0.

• potentials.txt: The number of synonymous and non-synonymous sites for each gene (determined by the gene’s codon composition). These values represent the denominators used to normalize pN and pS. A gene’s potential depends on which codons it contains: for example, a methionine codon (ATG) contributes zero synonymous sites because any single-nucleotide change to it is non-synonymous, whereas a leucine codon (CTN) contributes more synonymous sites. The potentials also depend on the --comparison parameter, since the comparison codon determines what is considered synonymous or non-synonymous.

Setting the right parameters

Please keep in mind that anvi-get-pn-ps-ratio has a few filtering choices that are set by default and may become critical for your investigation, including,

• The minimum departure from consensus for a variable position (--min-departure-from-consensus).
• The minimum departure from reference for a variable position (--min-departure-from-reference).
• The minimum number of SCVs in a grouping (--minimum-num-variants).
• The minimum coverage at a variable position (--min-coverage).

Please consider finetuning these parameters according to your research question.

Edit this file to update this information.

Additional Resources

Are you aware of resources that may help users better understand the utility of this program? Please feel free to edit this file on GitHub. If you are not sure how to do that, find the __resources__ tag in this file to see an example.