Add reference sequences for SARS-CoV-2

[donkirkby]

Support SARS-CoV-2 samples by adding the seed reference and other details.

• Add seed reference.
• Add coordinate references for gene regions, plus key positions if needed.
• Add landmarks for genome coverage map, based on gene regions.
• Handle duplicated base 13468.
• Add conseq_all.csv with a minimum coverage of 1, and only MAX cutoff.
• Compare consensus sequences with published results for the same samples.
• Handle duplicated base when aligning deletions. Split to Align deletions after frame shift #558.

[ArtPoon]

I already did this for the BC CDC - want the JSON?

[donkirkby]

Thanks, Art. Much appreciated!

[donkirkby]

Related conversation at PoonLab#42.

[rhliang]

Note to self: look at

• line 827 of aln2counts.py (this is the get_consensus_rows method)
• line 373 (where it calls self.write_consensus)

[donkirkby]

List of samples to download and the toolkit to download with.

Find more samples from SRA by searching for "Severe acute respiratory syndrome-related coronavirus"[orgn:__txid694009]. You can filter by platform, and there are currently 466 Illumina records.

[donkirkby]

Comparing consensus sequences to published results for my first 12 samples:

SRA accession	GISAID accession
SRR10903401
SRR10903402
SRR11092056
SRR11092057
SRR11092058
SRR11092064
SRR11140744
SRR11140746	EPI_ISL_408670
SRR11140748
SRR11140750
SRR11177792
SRR11314339

[ArtPoon]

You can see our work in progress here (with SRR numbers matched to published accession numbers):
https://github.com/PoonLab/sam2conseq/wiki

I'd be really curious to see what you get!

[donkirkby]

Thanks, @ArtPoon, that will save me a bunch of time. I'll let you know what we find.

[donkirkby]

Here's a summary of our results compared to the matches identified by @ArtPoon. (How did you find those, Art? Any suggestions on where to find EPI_ISL_408670? It doesn't show up when I search NCBI.)

Run	Compared to	Differences
SRR10903401-SARS_S1	MN988669.1	Very good: 12 mismatches in the first 24 bases under low coverage, and 21 extra A's at the end out of 29881.
SRR10903402-SARS_S2	MN988668.1	Almost perfect: 21 extra A's at the end out of 29881.
SRR11092056-SARS_S3	MN996530	Bad: 899 mismatches, 17761 missing, and 217 added out of 29854.
SRR11092057-SARS_S4	MN996528.1	Very good: 4 mismatches, 33 missing, and 12 added out of 29891. Missing 14 at the start, a gap of 15 with no coverage at 5397, plus 4 single gaps of no coverage within 20 bases. The mismatches are all in low coverage, 3 are mixtures when coverage is 2. 12 extra A's at the end..
SRR11092058-SARS_S5	MN996527.1	Bad: lots of sections with no coverage. 38 mismatches, 7606 missing, and 26 added out of 29825.
SRR11092064-SARS_S6	MN996531.1	Bad: lots of sections with no coverage. 24 mismatches, 4667 missing, and 33 added out of 29857.
SRR11140744-SARS_S7	EPI_ISL_408670	Almost perfect: 28 missing from the start, and poly-A tail replaced with ACAGATATATACGCC out of 29879.
SRR11140746-SARS_S8	EPI_ISL_408670	Almost perfect: poly-A tail replaced with AATAWMAACAAACAGAGCCTAAAAAGGACAAAA4 out of 29879.
SRR11140748-SARS_S9	EPI_ISL_408670	Almost perfect: 6 missing from poly-A tail out of 29879.
SRR11140750-SARS_S10	EPI_ISL_408670	Almost perfect: 9 missing from the start, and poly-A tail replaced with ACAATTGCAACAATC out of 29879.
SRR11177792-SARS_S11	MT072688	Almost perfect: 57 added out of 29811. A few added to start, most added at end: AGTGCTGAG + poly-A tail.
SRR11314339-SARS_S12	MT192765	Almost perfect: 38 added out of 29829. A few added to start, most added at end: CCATGTGATTTTAATAG + poly-A tail.

[donkirkby]

I forgot that I already had EPI_ISL_408670 downloaded. It came from GISAID. (Click on Browse, and search by accession id.)

[ArtPoon]

I queried the SRR number in the NCBI SRA database to get the sample description and then searched for a similar description in the GISAID annotations. Not perfect, I know.

[donkirkby]

Closing issue, now that basic reporting is working.