MyRT: identification and classification of reverse transcriptases in bacterial genomes and metagenomes

To make sure myRT pipeline works properly please follow these steps:

All the files needed for executing the pipelines are included in this github repository except cdd-pfamA.hmm files:

Download cdd-pfamA.hmm files

Use one of these commands to download the hmm files:

wget https://omics.sice.indiana.edu/myRT/Models/cdd-pfamA.tar.gz -P Models/

curl -L https://omics.sice.indiana.edu/myRT/Models/cdd-pfamA.tar.gz --output Models/cdd-pfamA.tar.gz

wget https://omics.informatics.indiana.edu/myRT/share/cdd-pfamA.tar.gz -P Models/

To decompress:

tar xvzf Models/cdd-pfamA.tar.gz -C Models/

Add read/execute permission if needed

chmod -R 777 Models/cdd-pfamA/ OR chmod a+rx Models/cdd-pfamA/ -R

After Successful uncompress delete the Zip file: rm -f Models/cdd-pfamA.tar.gz

Check the file permissions in myRT folder:

Use pwd to make sure you're in the main folder (myRT).

Make sure that you have permission to read and execute files in this folder.

You can use chmod -R 777 * or chmod a+rx * -R to add permission for all.

Install FragGeneScan

To install FGS Run these commands

current=$(pwd)

cd bin/FragGeneScan*

make clean

make fgs

cd ${current}

Test the Executable Files (if needed)

You can try these commands one by one to make sure all the executable files work

Rscript

bin/pplacer-Linux-v1.1.alpha19/pplacer

bin/FragGeneScan1.31/run_FragGeneScan.pl

bin/blast+/bin/blastp

bin/hmmer3.2/src/hmmscan

If hmmscan doesn't work, download the latest version of hmmer into bin folder and follow their installation instructions.

If you install/use a new version of these into bin folder, make sure you update the folder name in myRT-local-py3.py too.

Test Jplace Parser Rscript

Try this command to see if Jplace Parser (Scripts/JplaceParser.R) works

Rscript Scripts/JplaceParser.R QUERY/CP016020.1/CP016020.1-combo.jplace

If you already have treeio and castor package, you don't need to install them.

Please install a package only if you get an error that the package wasn't found.

If you need to install treeio then use the commands below, and modify Scripts/JplaceParser.R to include the path to the library:

mkdir ~/r-packages
R CMD INSTALL --library=~/r-packages bin/treeio_1.14.0.tar.gz
R CMD INSTALL --library=~/r-packages bin/castor_1.6.4.tar.gz

Next, modify the Scripts/JplaceParser.R use these packagee, for example:

library("treeio",lib.loc="~/r-packages")
library("castor",lib.loc="~/r-packages")

Then, try running JplaceParser.R again:

Rscript Scripts/JplaceParser.R QUERY/CP016020.1/CP016020.1-combo.jplace

You can check myRT-R.log to see what the output of Jplaceparser looks like.

You should also see this line if the jplace file is successfully parsed:

CP016020.1_3624763_3626067_+ RVT-UG3 1.00 {1887} leaf

This shows RT gene, prediction, confidence, {edgenum}, leaf/non-leaf

The result of parsing the jplace file will be saved in QUERY/CP016020.1/CP016020.1-pred.txt.

These files only exist if phylogenetic information were needed for a more accurate prediction, otherwise, these steps will be skipped.

You can modify the script to enforce the phylogenetic analysis, but the default is to skip if not needed.

Run myRT

Execute this command

./rerun.sh CP008696.1 QUERY

Make sure there's no "/" after QUERY.

Check the log file in myRT folder to see what to expect from these two commands.

Check the output files

If the run is successful you should have these two files, along with other files in QUERY/CP008696.1 folder:

ls QUERY/CP008696.1/CP008696.1-gn.gff

ls QUERY/CP008696.1/CP008696.1-RT.gff

You can compare it with QUERY/CP016020.1 folder if you want to make sure you have all the output files.

QUERY/BA000019.2 has less files in this folder compared to CP016020.1, because for this query the pplacer step was skipped (See Figure1 of our manuscript for more info).

See QUERY/README for a description of output files.

Download a reference genome from ncbi (if needed)

When all the dependencies are installed, try this command in the terminal to download this reference genome from ncbi:

./Scripts/download_ncbi.sh CP008696.1

If successful you should have a new folder (CP008696.1) in the QUERY folder.

To download to a new folder (use any name you'd like):

mkdir Newfolder
./Scripts/download_ncbi.sh CP008696.1  Newfolder

If successful you should have a new folder (CP008696.1) in the Newfolder folder.

If you'd like to download a list of reference genomes prepare a file like list_genomes and run the following command:

list_genomes should contain one accession number per line.

For example, CP008696.1 in line 1 and CP016020.1 in line 2 with no empty lines,and no space after accession number.

cat list_genomes | while read line ; do ./Scripts/download_ncbi.sh ${line} ; done

Other materials

The training data:

https://omics.informatics.indiana.edu/myRT/share/download.php

List of RT classes and compiled collection of RTs:

myRT-collection: https://omics.sice.indiana.edu/myRT/collection.php