Mismatch between mutant gene and reference gene outside the site of mutation

Question

I am tasked with designing primers for a particular mutant (Target Gene Locus: At1g28490 i.e. SYP61) of Arabidopsis, obtained from Gabi-Kat:

• The NCBI reference sequence for the genome goes: GGCTTTATAAGATTCTCCAGTAGCCCAAAAATAAACTTTCAGCGAATCTCAAAGCCAACCGCAGGTGATGCGTTTTGCTATGTAAGAGAGACGAATCTCACGCCGGAATCTCTTTTACTTCACCGTCGTTTGTTCTCCGTCGCGTCACCTTTCGATTAATTCCTTCACAGGGTAGCAAAATTTTGTTGATTTCGGG$\color{red}{\textrm{[}}$ATGTCTTCAGCTCAAGATCCATTCTACATTGTTAAAGAAGAGATTCAGGATTCTATTGATAAGTTGCAATCTACATTCCACAAATGGGAGCGTATCTCTCCCGACATGGGTGATCAAGCACATGTGGCGAAGGAGCTTGTTGCTACTTGTGGAAGCATTGAGTGGCAGGTA͟G͟A͟T͟G͟A͟G͟C͟T͟G͟G͟A͟A͟A͟A͟A͟G͟C͟G͟A͟T͟T͟A͟C͟T͟G͟T͟T͟G͟C͟A͟G͟C͟A͟A͟A͟A͟G͟A͟T͟C͟C͟T͟T͟C͟ATGGTATGGCATTGATGAAGCTGAGCTTGAAAAACGGAGGAGATGGACTAGTAATGCTAGGACACAGGTGCGGAATGTGAAGTCTGGTGTTTTAGCTGGAAAGGTTAGTTCTGGAGCAGGTCATGCAAGTGAAGTGCGTAGAGAACTAATGAGAATGCCAAACTCGGGTGAAGCAAGTAGATATGATCAGTATGGAGGGAGAGACGATGATGGTTTTGTACAGTCAGAATCAGATAGGCAAATGCTGTTGATAAAGCAACAAGACGAAGAATTGGATGAGCTTAGTAAAAGTGTACAGAGAATTGGAGGAGTGGGGCTCACTATACACGACGAACTCGTTGCACAGGAGAGAATAATAGATGAATTGGATACGGAGATGGACAGCACAAAGAACCGTCTTGAATTTGTTCAGAAAAAAGTGGGAATGGTAATGAAGAAAGCTGGAGCAAAAGGTCAAATGATGATGATATGTTTCTTGCTCGTCTTGTTCATCATCCTATTCGTTCTTGTCTTCTTGACCTAA$\color{red}{\textrm{]}}$AGAACAGAGCCGCGGGATCTTATAGATGGGAGTTTAATGTCTTACATTATCATGTTTGTGGTGAAACAAATGGAATATTTACATAAGGTTGTCTGTTTTTTCTTCTCCTATCCAATCCCACATATAATTTTGAACTTTGGCTGTGTGATCTCTAGAGAACCTGAAATTAAATACTTTCATAGCTTGATCCCATGGCTAAGTTCCT
• Gabi-Kat provides the following sequence from the mutant as a confirmatory match with the genome locus: AGCGGATCTCTCCCGACATGGCTGATCAAGCACATGTGGCGAAGGAGCTTGTTGCTACTTGTGGAAGCATTGAGTGGCAGGTC͟C͟T͟C͟C͟T͟A͟T͟T͟C͟A͟C͟C͟C͟A͟C͟G͟G͟T͟T͟C͟T͟A͟T͟T͟A͟A͟T͟C͟C͟C͟C͟T͟C͟T͟T͟T͟A͟T͟A͟T͟A͟T͟A͟A͟

I have bold-faced the common region to both sequences. What explains the complete mismatch of the trailing part (u͟n͟d͟e͟r͟l͟i͟n͟e͟d͟) of the Gabi-Kat sequence with the ref genome?

It is not the case that the mutation happened there; we are in the coding region (demarcated by red square brackets in the NCBI sequence) while Gabi-Kat's mutation is in an untranslated region.

Update 1:

Running Ensembl Blast on the GabiKat seq returns a 92.9% identicality with SYP61. Their sequence for the gene (taken from TAIR10.1) goes:

AGCGGATCTCTCCCGACATGGCTGATCAAGCACATGTGGCGAAGGAGCTTGTTGCTACTTGTGGAAGCATTGAGTGGCAGGTA͟C͟T͟C͟G͟T͟A͟T͟T͟T͟A͟G͟T͟C͟A͟C͟G͟G͟T͟T͟C͟T͟A͟T͟T͟A͟A͟T͟C͟T͟C͟G͟T͟C͟T͟T͟T͟A͟T͟A͟T͟A͟T͟A͟A͟

Note the difference in the trailing part from NCBI ref; this almost coincides with the Gabi-Kat seq!

So, is the NCBI ref-genome and the data for the particular TAIR genome inaccurate? Maybe, TAIR updated their whole genome set but forgot to update data for individual genes?

Update 2:

My proposed explanation for the anomaly makes no sense. TAIR 7 (2004) has the same sequence. It looks like I am missing out on something more fundamental.

Answer 1

Conclusion The evidence suggests the region you are looking at Gabi-kat appears either 27045 downstream or 192250 upstream of syntaxin SNARE (SYP61), but is definitely not within the gene. It is chance homology (e.g. a pseudo-gene) that appears to occur within syntaxin (SYP61). The intergenic region could be much closer to the SYP61 gene if the same reference genome was used for both analyses Genbank LR782542.1 versus Genbank NC_003070.9 (different reference genomes in my analysis). Hence, the University is probably correct.

What I suggest is the genome of A. thaliana is downloaded and Blast analysis is performed locally. This way differences in the reference genome are removed.

The analysis is as follows:

The target protein in question is involved in vesicle trafficking (here) called SNARE, comprising coiled-coil domains and are highly conserved, for example show 100% conservation between Arabidopsis thaliana, Arabidopsis suecica and Camelina sativa

MSSAQDPFYIVKEEIQDSIDKLQSTFHKWERISPDMGDQAHVAKELVATCGSIEWQVDELEKAITVAAKDP

If Gabi-kat was a protein its sequence would be ...

RISPDMADQAHVAKELVATCGSIEWQVLLFTHGSINPLFIY

Here's a blast of the whole regions of the Gabi-kat. The region in green is the SNARE proteins ...

So the issue is what is the terminal 3' protein and Blast that, i.e. the final bit that matches no known plant and the locus you flagged as being anomalous ..

This is basically zero result via Blast, the protein with low homology are lower metazoans (animals) and a few bacteria ... a long way from plants. In other words this isn't a known protein of natural origin at least on NCBI.

The locus is literally like no protein on earth.

However ... Blasting the nucleotide ...

1. Reference Arabidopsis thaliana = syntaxin (SNARE; SYP61)

2. Gabi-Kat Arabidopsis thaliana top two Blast hits are ...

• Chromosome 1 at 10039082 - 10039208, or 9819787 to 981991 or
• Both value 1e-42
• Both dentity is 93%

So Gabi-Kat is a real part of the Arabidopsis thaliana, Gabi-kat is a non-coding region. Thats the major conclusion.

The question where is the location of SYP61? And that is

• Chromosome 1 at 10012037 to 10020311

Described here from NCBI's full report

Thus the mutant is 27045bp downstream or 192250 bp upstream IF the same reference genome is used for the blasting (Genome LR782542.1) and the Genbank report NC_003070.9. They could easily be much closer to each other is the same reference genome was used for both sets of analysis.

---

To answer the comments

• I [the OP] do not follow your [my] blast analysis
• My [the OP] blasts gives ... ['hits within the gene']
• etc ....

It appears to me there is a psuedo-gene phenomenon occurring which is making you think a mutant is located within a gene. However, if you disagree thats perfectly okay and the original version of this answer I had assumed the Gaba-kat team had got this wrong (I always from the basis of supporting the OP).

NCBI was the database repository in the remote analysis, but there's an issue with multiple references ... so to reiterate I suggest you perform this analysis as a local blast to remove multiple reference phenomena.

To conclude I personally don't think there is a basis to believe the University responsible for Gaba-kat has got this wrong https://www.gabi-kat.de. However, if you feel otherwise then I would recommend contacting them with your analysis and any supportive analysis for example from this thread or elsewhere. Please note, I don't have a connection with this university.