Add CHAMPS pathogen reference genomes to chromosomes.tsv

[Copilot]

Adds complete-genome INSDC accessions for pathogens in the CHAMPS child mortality surveillance panel that were missing from Kalamari, with a preference for accessions from NCBI-designated reference genomes.

New entries

• Bacteria: Actinobacillus pleuropneumoniae, Aeromonas hydrophila, Bordetella holmesii/parapertussis, Brucella abortus/melitensis (2 chromosomes each), Corynebacterium ulcerans, Mycoplasma pneumoniae, Orientia tsutsugamushi, Salmonella Typhi/Paratyphi A, Treponema pallidum, Ureaplasma urealyticum
• Single-segment viruses: CMV, HSV-1/2, VZV, HCoV-229E/NL63/OC43/HKU1, MERS-CoV, SARS-CoV-1, EV-A71, Rhinovirus A, Parechovirus 1, hMPV, HPIV 1–4, Mumps, Nipah, RSV-A, DENV 1–4, JEV, WNV, YFV, Zika, CHIKV, Rubella, HEV, HAdV-2/40/41, HAstV-1, Norovirus GI/GII, Sapovirus GI, Parvovirus B19
• Multi-segment viruses: Influenza A H1N1 pdm09 (8 segments: CY121673–CY121680), CCHFV (3), Lassa (2), RVFV (3), Rotavirus A SA11-H96 (11 segments: EF672557–EF672567)
• Fungi: Candida albicans SC5314 (8 chromosomes: CP017623–CP017630)

Accession selection

All accessions are INSDC (GenBank), not RefSeq. Where an NCBI-designated reference genome exists, its INSDC source accession is used:

Organism	Accession	Notes
SARS-CoV-1	AY274119	NC_004718 source (Tor2/BCCA); replaces BJ01 isolate AY278491
Dengue virus 3	M93130	NC_001475 source (H87 prototype)
JEV	AF217620	NCBI GenBank complete genome; D90194 (Nakayama/DDBJ) excluded — not retrievable via efetch
RSV-A	AF013254	NC_001781 source (complete Long strain); replaces shorter M74568
Rotavirus A	EF672557–EF672567	SA11-H96 strain, NC_011503–NC_011513 sources; replaces 11 disparate 1980s–2000s segments

Zika is represented by KX369547 (PRVABC59, 2015 epidemic strain) rather than AY632535 (1947 mouse-adapted MR766), which is more relevant for surveillance.

Also fixed

• Corynebacterium ulcerans taxid corrected: strain-level 38289 → species-level 1714

Original prompt

---

This section details on the original issue you should resolve

<issue_title>CHAMPS organisms</issue_title>
<issue_description>There is a multinational collaboration called CHAMPS where they test samples from deaths of very young or stillborn people. Experts determine cause of death, and samples are tested postmordem to identify the pathogens. There is a strong ability to link cause of death to individual pathogens. There is also a test for more than 140 pathogens. I think it would be great to get the reference genomes of these pathogens and compare against CHAMPS results in the future.

Code	Assay	Respiratory V1/V2	Respiratory V3	Enteric	Tier 1 (Blood/CSF)	Tier 2 (Blood/CSF)
ABAU_1	Acinetobacter baumanii	X	X			X
ADEV_1	Adenovirus	X	X	X	X
BPAR_1	Bordetella parapertussi, Bordetella bronchiseptica (Insertion sequence IS1001)	X	X
BOP2_1	Bordetella pertussis (pertussis toxin)	X	X
BOP1_1	Bordetella pertussis, Bordetella holmseii (Insertion sequence IS481)	X	X
BUPS_1	Burkholderia pseudomallei	X	X			X
CHTR_1	Chlamydia trachomatis	X	X
CODI_1	Corynebacterium diptheriae	X	X
COUP_1	Corynebacterium ulcerans, Corynebacterium pseudotuberculosis	X	X
CTOX_1	Corynebactrium spp. (tox gene)	X	X
CYMV_1	Cytomegalovirus (CMV)	X	X			X
ENTV_4	Enterovirus	X	X	X	X
GAST_2	Group A Streptococcus	X	X		X
GBST_1	Group B Streptococcus	X	X		X
HIAT_1	Haemophilus influenzae	X	X		X
HITB_2	Haemophilus influenzae type B	X	X		X
HCV1_1	Human coronavirus 1 (229E)	X	X
HCV2_1	Human coronavirus 2 (NL63)	X	X
HCV3_1	Human coronavirus 3 (OC43)	X	X
HCV4_1	Human coronavirus 4 (HKU1)	X	X
HMPV_1	Human metapneumovirus (HMPV)	X ...

• Fixes CHAMPS organisms #74

---

📱 Kick off Copilot coding agent tasks wherever you are with GitHub Mobile, available on iOS and Android.

[lskatz]

I'll have to check each of the accessions to see if it

• is a legitimate accession
• is on the CHAMPS list (or why is it important?)
• is a complete genome
• has a correct taxid
• has a correct parent taxid
• is a complete genome
• is recommended somehow (NCBI reference genome? CDC SME? Other SME?)

Checking taxonomy one by one (maybe there's a better way)

$ esearch -db nuccore -query CP000462 | elink -target taxonomy | efetch -format xml | xtract -format | grep -EC 4 "<Rank>genus|<Rank>species"
      </Taxon>
      <Taxon>
        <TaxId>642</TaxId>
        <ScientificName>Aeromonas</ScientificName>
        <Rank>genus</Rank>
      </Taxon>
      <Taxon>
        <TaxId>644</TaxId>
        <ScientificName>Aeromonas hydrophila</ScientificName>
        <Rank>species</Rank>
      </Taxon>
      <Taxon>
        <TaxId>196023</TaxId>
        <ScientificName>Aeromonas hydrophila subsp. hydrophila</ScientificName>

[lskatz]

@copilot can you suggest other edirect commands to verify contig completeness, taxid, parent taxid, and inclusion in the NCBI reference genomes list?

[Copilot]

@copilot can you suggest other edirect commands to verify contig completeness, taxid, parent taxid, and inclusion in the NCBI reference genomes list?

Here are edirect commands for each verification type. The key fields from efetch -format docsum on the nuccore database are TaxId, Length, Completeness, and Genome (chromosome/complete/plasmid/etc).

1. Verify accession, taxid, completeness in one shot:

acc="AF013254"
esearch -db nuccore -query "${acc}[accn]" | efetch -format docsum | \
  xtract -pattern DocumentSummary -element Caption TaxId Length Completeness Genome Organism

2. Verify taxid and get parent taxid:

acc="AF013254"
esearch -db nuccore -query "${acc}[accn]" | elink -target taxonomy | \
  efetch -format xml | \
  xtract -pattern Taxon -element TaxId ScientificName -block LineageEx/Taxon -tab "\n" -element TaxId ScientificName Rank

The parent taxid is the last entry before the species in the lineage.

3. Check if it's an NCBI reference genome (RefSeq_category = "reference genome"):

acc="AF013254"
esearch -db nuccore -query "${acc}[accn]" | elink -target assembly | \
  efetch -format docsum | \
  xtract -pattern DocumentSummary -element AssemblyAccession AssemblyStatus RefSeq_category Organism

If nothing comes back, the accession has no assembly record (common for viral sequences). For those, check manually at https://www.ncbi.nlm.nih.gov/datasets/genome/?taxon=11250

4. Batch-check all CHAMPS accessions from chromosomes.tsv:

tail -n +2 src/chromosomes.tsv | awk -F'\t' '{print $1"\t"$2"\t"$3"\t"$4}' | \
while IFS=$'\t' read name acc taxid parent; do
  result=$(esearch -db nuccore -query "${acc}[accn]" | efetch -format docsum | \
    xtract -pattern DocumentSummary -element Caption TaxId Length Completeness)
  echo -e "$name\t$acc\t$taxid\t$parent\t$result"
done

5. Completeness field values to look for: Complete means it is a complete genome. Chromosome also indicates assembled-to-chromosome level. Contig or empty means it may be incomplete.

6. Verify the NCBI RefSeq reference genome for a given taxid:

taxid="11250"  # RSV-A
esearch -db assembly -query "txid${taxid}[Organism:exp] AND \"reference genome\"[RefSeq Category]" | \
  efetch -format docsum | \
  xtract -pattern DocumentSummary -element AssemblyAccession SpeciesName RefSeq_category

This returns the RefSeq assembly record; you can then trace back to the INSDC accession from there.