anvi-get-sequences-for-hmm-hits

Get sequences for HMM hits from many inputs.

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

Authors

A. Murat Eren (Meren)

WebEmailTwitterGithub

Can consume

contigs-db profile-db external-genomes internal-genomes hmm-source hmm-hits

Can provide

genes-fasta concatenated-gene-alignment-fasta

Usage

This program can work with anvi’o contigs-db, external-genomes, or internal-genomes files to return sequences for HMM hits identified through the default anvi’o hmm-sources (such as the domain-specific single-copy core genes) or user-defined hmm-sources (such as HMMs for specific antibiotic resistance gene families or any other targets).

Using it with single-copy core genes in default anvi’o HMMs make it a very versatile tool for phylogenomics as the user can define specific sets of genes to be aligned and concatenated.

Learn available HMM sources

Get all sequences in a given HMM source

Learn available genes in a given HMM source

Please note that the flag --list-available-gene-names will give you the list of genes in an HMM collection (for example, for Bacteria_71 in the following use case), and it will not give you the list of genes in your genomes or metagenomes that are matching to them. You can generate a table of HMMs across your genomes or metagenomes with another program, anvi-script-gen-hmm-hits-matrix-across-genomes.

Get sequences for some sequences in a given HMM source

Get HMM hits in bins of a collection

Get amino acid sequences for HMM hits

Get HMM hits independently aligned and concatenated

The resulting file can be used for phylogenomics analyses via anvi-gen-phylogenomic-tree or through more sophisticated tools for curating alignments and computing trees.

Please note teh presence of a new flag in this particular command line, --return-best-hit. This flag is most appropriate if one wishes to perform phylogenomic analyses, which ensures that for any given protein family, there will be only one gene reported from a given genome. This is necessary due to the nature of the data that goes into phylogenomic analyses, where typically multiple single-copy core genes from each genome are individually aligned and the results are concatenated into a super matrix for tree construction. This requirement will be violated if for a given single-copy core gene (SCG) family any given genome in the dataset has two or more genes rather than one, which can happen for a variety of technical or biological reasons. In that case, we need to pick only one of those genes, which is exactly what --return-best-hit flag does for us. Let’s say we have two Ribosomal_L3 gene hits in a given genome. When declared, this flag will choose the Ribosomal_L3 gene that has the most significant hit given the hidden Markov model for Ribosomal_L3 that was used to search for Ribosomal_L3 genes in genomes. In cases where genome quality is sufficient and contamination is not a considerable risk, this step will choose the right hit as in many cases of multiple hits for SCGs the additional ones will have very low significance. Essentially, --return-best-hit makes sure you are working with the most appropriate genes for phylogenomics given the HMM modesl and significance scores for your matches in your genomes.

Tips

Get amino acid seqeunces for each gene in a model individually

If you are interested in recovering HMM hits for each gene in a model anvi’o knows about as a separate FASTA file, you can do it with a for loop easily. After learning your genes of interest, first run this to make sure your terminal environment knows about them (this is an example with a few genes from the HMM source Bacteria_71, but you can add as many genes as you like and use any HMM source anvi’o recognizes, of course):

export genes="Ribosomal_L22 Ribosomal_L23 Ribosomal_L27 Ribosomal_L27A Ribosomal_L28"
export hmm_source="Bacteria_71"

Then, you can run the program in a loop to have your FASTA files:

for gene in $genes
do
    anvi-get-sequences-for-hmm-hits -c CONTIGS.db \
                                    --hmm-source $hmm_source \
                                    --gene-name $gene \
                                    -o ${hmm_source}-${gene}.fa
done

Voila!

Exercise with the program or test scenarios

You can play with this program using the anvi’o data pack for the infant gut data and by replacing the parameters above with appropriate ones in the following commands.

Download the latest version of the data from here: doi:10.6084/m9.figshare.3502445

Then, unpack it:

Finally, import the collection merens:

Then run the program using the PROFILE.db, CONTIGS.db, and optionally the collection merens to try some of the commands shown on this page.

Edit this file to update this information.

Additional Resources

• A tutorial on anvi'o phylogenomics workflow

• A detailed application of phylogenomics to place a new genome on a tree

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.