Whole Genome Sequencing for Rapid Characterization of Rabies Virus Using Nanopore Technology

Our research uses advanced genomics to enhance pathogen surveillance in low and middle-income countries. And this helps us empower local researchers to answer five big questions, who, what, where, why, and when. And this can inform practical solutions on the ground.

We use nanopore technology to decentralize sequencing. Its ease of use, portability, and cheaper costs relative to other platforms makes it ideal for low-resource settings. And the ability to rapidly sequence at the sample source means that we can do surveillance at the animal-human-environment interface.

We face challenges in obtaining whole genomes from samples with varying quality. Often, submissions from remote areas would be transported in poor storage conditions. And also, we have limited expertise in buying informatic tools that would be useful in interpreting data.

So these tools would be like MADDOG and RABV-GLUE. This could be used in interpreting data in a universal way to help initiate control. Okay, we are trying to demonstrate how to optimize the protocol for rapid sequences for whole genomic sequences of rabies and to see how it can be adapted and be used in the low-income countries for sequencing of rabies.

So as to answer different questions on how it really is circulating, what is circulating, and when is it circulating. And this information's necessary for rabies elimination strategies. Viral genomic surveillance is now more accessible.

However, guidance for data analysis and interpretation for local control efforts is more limited. Our protocol offers an end-to-end pipeline, empowering local scientists with frontline surveillance capabilities. Greater availability of viral genomes and methods for interpreting this data allows researchers to translate epidemiological insight into actionable information for government ministries or disease control programs, and guide the decision-making process.

Prepare two ceramic beads tubes by filling a two-milliliter PCR tube with an approximately 200-microliter tube full of 1.4-millimeter ceramic beads and label the tube accordingly before transferring them inside the sterile hood. Add the recommended volume of lysis buffer provided in the RNA extraction kit to the labeled PCR tube. Get approximately a three-millimeter cube from the brain sample confirmed with rabies infection using a wooden applicator and put it into a labeled tube with sample ID and 100 microliters of nuclease-free water into the tube labeled negative control.

Disrupt the brain tissue manually using a wooden applicator stick, and then vortex at maximum speed until complete tissue homogenization is achieved. Next, centrifuge the homogenized lysate. Transfer the supernatant to a newly-labeled microcentrifuge tube, using a pipette, and use it for further RNA extraction steps, CDNA preparation, and PCR amplification.

After the CDNA preparation and PCR amplification using primer pools A and B, perform PCR cleanup and quantification in the post-PCR area. Aliquot SPRI beads into microcentrifuge tubes from the main bottle. These can be also prepared in advance.

Store the tubes at four degrees Celsius or keep them in a cold rack or ice. Next, warm the SPRI bead aliquot at approximately 20 degrees Celsius and thoroughly vortex to resuspend the beads in the entire solution. Then, in 1.5-milliliter tubes, combine the primer pool A and primer pool B PCR products for each sample.

Add water to bring the volume to 25 microliters if required before adding 25 microliters of SPRI beads to each combined product and mixing it. Incubate the mixture at room temperature for 10 minutes with occasional inverting or flicking of the tubes. Next, place the tubes on a magnetic rack until the separation of beads and solution.

Once done, discard the supernatant without disturbing the bead pellet. Then, give two washes of 30 seconds using freshly-prepared 200 microliters of 80%ethanol and discard the ethanol. After the second wash, remove all traces of ethanol using a 10-microliter pipette tip.

Air dry the pellet until trace ethanol has evaporated and the pellet changes from shiny to matte. Resuspend the beads in 15 microliters of nuclease-free water to recover the cleaned DNA product and incubate at room temperature for 10 minutes. Separate the beads from the solution on a magnetic rack.

After transferring the supernatant to a fresh 1.5-milliliter tube, prepare a 1:10 dilution of each sample in nuclease-free water. Measure the DNA concentration of each diluted sample with a highly-sensitive and specific fluorometer, following the manufacturer's instructions. Begin the priming and loading the quality-checked flow cell by flipping back the sequencing device lid and sliding the priming port cover clockwise, visualizing the priming port.

To remove air bubbles, set a P1000 pipette to 200 microliters and insert the pipette tip vertically into the priming port. Turn the wheel until a small volume entering the pipette tip is seen. Load 800 microliters of pre-prepared flow cell priming mix into the flow cell via the priming port to avoid introducing bubbles.

Lift the sample port cover and load 200 microliters of the remaining priming mix into the flow cell via the priming port. To ensure the mixing of the loading beads, resuspend the library master mix by pipetting and dropwise, load 75 microliters to the flow cell via the sample port. Replace the sample port cover gently, ensuring the bung enters the sample port.

Close the priming port and replace the sequencing device lid. For live base calling, use RAMPART. Use the artic-rabv environment and work in the directory created for the Rampart_Output.

Then, type the RAMPART command to navigate to the required paths. First, the RAMPART-specific scheme protocol, and next, basecalledPath, the minknow fastq_pass output folder for the run. Open a browser window and navigate to localhost:3000 in the URL box.

Wait for sufficient data to be base called before results appear on the screen. The top three panels show summary plots for the whole run. Plot one shows the depth of coverage of mapped reads for each barcode per nucleotide position on the index reference genome.

Plot two shows mapped reads from all barcodes over time. And plot three shows mapped reads per barcode. Lower panels show rows of plots per barcode.

The left shows the depth of coverage of mapped reads per nucleotide position on the index reference genome. The length distribution of mapped reads is in the middle. The proportion of nucleotide positions on the index reference genome obtaining 10X, 100X, and 1, 000X coverage of mapped reads over time is seen in the right corner.

For lineage assignment of consensus sequences, use MADDOG. Pull the MADDOG repository from GitHub to ensure working with the latest version. Create a folder within the previously-created local MADDOG repository.

Inside the folder, add the FASTA file containing the consensus sequences. Also, add a metadata file to the folder. Ensure that this file is a CSV with four columns called ID, country, year, and assignment.

Pull the MADDOG repository from GitHub to ensure working with the latest version. In the command line interface, activate the conda environment with the conda activate MADDOG command. In the command line interface, navigate to the MADDOG repository folder.

First, perform lineage assignment on sequences to check for potential abnormalities and identify if running the longer lineage designation step is appropriate by running the sh assignment. sh command. When prompted, enter Y to confirm that the repository is pulled and working with MADDOG's latest version.

When prompted, enter the folder name containing the FASTA file within the MADDOG repository. When the lineage assignment is complete, check the output file in the folder. If the output is as expected and multiple sequences are assigned to the same lineage, then run the lineage designation.

While running lineage designation, delete the assignment output file just created. In the terminal, inside the MADDOG repository folder, run the command sh designation.sh. When prompted, enter Y to indicate that the repository is pulled and the work is being done with the most up-to-date version of MADDOG.

When prompted, enter the folder name within the MADDOG repository folder containing the FASTA file and metadata. The sample-to-sequence-to-interpretation workflow for rabies virus RABV was successfully used in different laboratory conditions in endemic countries, such as Tanzania, Kenya, Nigeria, and the Philippines. The live base calling using RAMPART showed the real-time generation of reads and the percent coverage per sample.

A lineage classification in nomenclature system, MADDOG, used to compile and interpret resulting RABV sequences showed the higher resolution classification of local lineages, following the MADDOG assignment.