So you want to do genotype work on Newcastle disease virus samples

It starts with pulling RNA out of a tracheal swab or lung tissue and running RT-PCR to amplify the fusion gene. That is the standard region everybody looks at. The primer sets are mostly based on the WHO/OIE reference panels from the last decade. You get your amplicon, clean it up, send it to a sequencing facility, and then you are staring at a chromatogram trying to figure out whether that dip in the trace is a real mixed infection or just your machine being lazy. I spent about three years in a diagnostic lab doing exactly this. The problem most people hit first is that NDV field strains often carry insertions and deletions in the fusion protein cleavage site. The standard primers skip over some of that variability, so you end up with poor coverage in the very region you care about for pathotyping. One time I had a sample that typed as genotype VII but the Sanger trace kept breaking down after cycle 180. Turned out there was a four-nucleotide insertion right near the primer binding site that I had missed because the reference sequence I was aligning against was from 2003. I redesigned the forward primer two bases further upstream, re-ran the PCR with a touch-up annealing temperature of 58 degrees instead of the usual 55, and got clean coverage through the whole cleavage site region. That fixed it.

Unveiling The Secret World Of Ndv Genotypes A Probing Analysis

The basic workflow is straightforward enough. Extract nucleic acid, run reverse transcription, amplify the ~1.7 kilobase fragment covering the F gene and part of the P gene, sequence it, and then build a phylogenetic tree. The hard part is the sequence quality and the tree building. You need to make sure your amplification is not biased toward one strain when you have a co-infection. NDV mixes are common in flocks that have been vaccinated and then exposed to field virus. If you run a single round of PCR and clone the product before sequencing, you will spend weeks picking colonies. Direct Sanger sequencing of the mixed product gives you a messy trace but you can usually read through the minority variant if you have enough template. I typically run two separate PCRs from the same cDNA and compare the chromatograms. When they disagree, I know I am dealing with a quasispecies situation and I move to next-generation sequencing if the question matters. The genotyping scheme most people use is the one from Afshar and elsewhere, updated through the VII cases. There are twelve major genotypes now, with VII and VIIe dominating outbreaks in Asia and Africa. Genotype II is still in commercial vaccines. When you are looking at a new isolate, you align it against a reference set and check the nucleotide identity at the F protein cleavage site. That site is what tells you whether the virus is velogenic, mesogenic, or lentogenic. The rule of thumb is six to eight amino acids in the motif, but there are edge cases. Some VIIe strains have a slightly atypical motif and still cause severe disease. You cannot rely on that alone. For the phylogenetic analysis, I use MAFFT for alignment and IQ-TREE for tree construction with 1000 bootstrap replicates. The model selection is automatic with the –m TEST flag. It is fast and usually gets the topology right for NDV. BEAST is overkill unless you are doing a molecular clock analysis for a publication. The runtime on a decent machine is about forty-five minutes for a typical dataset of fifty sequences. Download the reference alignment and primer sequences from the OIE NDV resource page if you do not have them locally. Most university virology labs keep a copy. The NCBI GenBank has the full set of reference sequences under accession numbers that start with the standard prefixes. You can also pull them from the Nextstrain NDV dataset if you want something that is already curated and time-stamped. One thing that trips people up is the naming. Genotype names shifted a few times as more sequences came in. A paper from 2018 might call something VII while a 2024 paper calls the same clade VIIe. Check the date of the nomenclature you are using. The current scheme is maintained by the WHO/OIE collaboration and gets updated when a new major clade emerges. If your isolate does not fit any existing genotype, you may be looking at a novel cluster, which is rare but happens. The main bottleneck is sample quality. Degraded RNA from field specimens gives weak amplification and incomplete reads. I always include an internal control gene in the RT step, usually the NP gene, to verify that the extraction worked. If the control amplifies but the F gene does not, the sample is either too degraded or the primers do not match. Redesigning primers for divergent strains is the usual fix, but it takes time. NGS is becoming more practical for this work. Illumina MiSeq runs with a custom amplicon panel can give you full-length genome sequences in a single run for about thirty dollars per sample if you multiplex twenty runs together. The downside is the bioinformatics. You need a pipeline to filter low-quality reads, assemble contigs, and call variants. I use Geneious for the assembly and write a small Python script to extract the F gene and annotate the cleavage site automatically. Without automation, the manual annotation of sixty sequences takes about six hours of tedious reading. If you are just starting out, stick to Sanger sequencing of the F gene fragment. It is cheaper, faster to process, and sufficient for most diagnostic and surveillance purposes. Move to NGS only when you need full genome data or when you are working with highly mixed populations where Sanger traces are unreadable. The extra effort is worth it for outbreak investigations but not for routine typing.