2024-07-14 snapshot stats and major changes

[pgaudet]

stats are here: https://docs.google.com/spreadsheets/d/1kTSYUTrgDXpXlXJpekG3ifRCbjjIqinnI3vH3DgU_BU/edit?gid=0#gid=0

[pgaudet]

Major changes:

BY ORGANISM Human: + 82870

• +18760 IDA
• +60678 IPI >> many new IntAct& ComplexPortal
• -9848 gene products
• Are these to isoforms?
  • 1466 "-PRO" in col 17
  • 45,000 have '-' (ie, isoforms or UniProt "chains")
  • New RNAs that are now not being converted to 'gene product':
    • pre_miRNA: +3257
    • lncRNA: +22082
    • rRNA: -6965

AmiGO versus AmiGO-staging:

• 707156 human annotations in AmiGO
• 769603 human annotations in AmiGO-staging
• diff is 62,447; not 82,870 as shown in the stats

Source files

• 2024-06-17 release:
  • source: 705016 lines (note that this includes headers but they are the same)
  • GOC-produced file: 707209 lines
  • 2024-07-14 snapshot:
  • source: 767735 lines
  • GOC-produced file: 769659 lines

So, I am not sure why the stats are inconsistent. But the data in the files looks right relative to the upstream. In conclusion, Human data looks OK

---

Mouse : + 67644

• +4768 IDA >> AmiGO versus staging: UniProt + ~ 1000 new mouse IDAs. But there is not 4700 new IDAs
• +3595 IPI >> many new IntAct& ComplexPortal
• +37703 ISO >> pipeline moved to GOC
• -13447 IEA

Mouse IEAs: AmiGO: 69359

• EC 1115
• InterPro 17944
• UniProtKB-KW 50300
• (all assigned_by MGI)

Staging: 55912

• Ensembl 26172
• UniProt 20516
• InterPro 5180
• GOC 2851
• RHEA 1193

Much better!

Drosophila: + 4427

• 1542 IDA
• 2500 IMP

• 145 IGI

  BY GROUP MGI : -138973 / 179719 | -77.33% HPA** : + 14103 / 24306 | 58.02%

[pgaudet]

CHANGES IN REFERENCES AND PMIDS BY GROUP GOC	-843 / 1000	-84.30%	-419 / 499	-83.97%
YuBioLab	-131 / 259	-50.58%	0 / 131	0.00%
NTNU_SB	-251 / 564	-44.50%	-12 / 531	-2.26%
DFLAT	-18 / 47	-38.30%	0 / 45	0.00%
ParkinsonsUK-UCL	-269 / 829	-32.45%	2 / 704	0.28%
SynGO-UCL	-9 / 33	-27.27%	-1 / 21	-4.76%
Alzheimers_University_of_Toronto	-15 / 57	-26.32%	1 / 48	2.08%
GOC-OWL	-14 / 56	-25.00%	-7 / 28	-25.00%
BHF-UCL	-1385 / 5616	-24.66%	16 / 5193	0.31%

---

CHANGES IN REFERENCES AND PMIDS BY TAXON taxon	references	% references	pmids	% pmids
NCBITaxon:9606|Homo sapiens	1277 / 59664	2.14%	1115 / 46960	2.37%
NCBITaxon:10116|Rattus norvegicus	38 / 49838	0.08%	18 / 24695	0.07%
NCBITaxon:10090|Mus musculus	-14473 / 36058	-40.14%	241 / 34595	0.70%

This is because internal MGI reference IDs have been removed. >> OK.

[pgaudet]

CHANGES IN ANNOTATED BIOENTITIES BY FILTERED TAXON AND BY BIOENTITY TYPE (ALL)   taxon	protein	% protein	gene_product	% gene_product
NCBITaxon:9031|Gallus gallus	14 / 25514	0.05%	-542 / 2	-27100.00%
NCBITaxon:9615|Canis lupus familiaris	324 / 33820	0.96%	-764 / 3	-25466.67%
NCBITaxon:9823|Sus scrofa	53 / 26619	0.20%	-1640 / 22	-7454.55%
NCBITaxon:9913|Bos taurus	6011 / 28331	21.22%	-808 / 25	-3232.00%
NCBITaxon:4896|Schizosaccharomyces pombe	1 / 5040	0.02%	-16 / 1	-1600.00%
NCBITaxon:9606|Homo sapiens	31 / 19752	0.16%	-9848 / 706	-1394.90%

This is because we fixed GORULE-0000001 and we now recognize more entities (for example lncRNAs are not converted to 'gene product' anymore). (see https://github.com/geneontology/go-site/issues/2246) >>> OK.

[pgaudet]

CHANGES IN ANNOTATIONS BY TAXON Taxon	Values	% change	reason
NCBITaxon:9606|Homo sapiens	62447 / 769603	8.11%
NCBITaxon:10090|Mus musculus	29005 / 552051	5.25%
NCBITaxon:9913|Bos taurus	22066 / 208681	10.57%
NCBITaxon:242507|Pyricularia oryzae 70-15	14170 / 14170	100.00%	PANGO ND!?!?
NCBITaxon:3218|Physcomitrium patens	33171 / 33171	100.00%	New IBAs
NCBITaxon:243230|Deinococcus radiodurans R1 = ATCC 13939 = DSM 20539	2811 / 2811	100.00%	New IBAs

[pgaudet]

• NEW REFERENCES: 2120 - to isoforms
• NEW PMID:1557
• • PMID: 38382089 >> TAIR, not in QuickGO
• REMOVED REFERENCES: 15,000 (14.740 MGI)

[pgaudet]

New isoform data displayed correctly on AmiGO staging:

This data was missing from AmiGO - great that we now have it!

[pgaudet]

OK everything looks fine!

[pgaudet]

Emailed Sandra to inquire about IntAct & ComplexPortal releases

[pgaudet]

@suzialeksander Major changes: (you can decide which points are worth mentioning):

• GOC now loads isoform data for human, dog, chicken, pig and cow. This results in about 45,000 new annotations for human only
• A bug has been fixed with respect to how some non-coding RNA are imported, resultsing in about 25,000 new annotations for lncRNA and pre-miRNAs (that used to be converted to the more generic 'gene product')
• MGI pipeline now ran at GO Central - we now have annotations from Ensembl, and InterPro, EC and RHEA IEAs are now displayed using these sources as 'contributors'
• MGI internal references have been removed from GO products.
• ComplexPortal, IntAct and Human Protein Atlas (HPA) has made a new release, increasing the number of protein-protein interactions (in particular for human, +60678, and for mouse, + 3595), and increasing by 14000 the number of human CC.