Hey @GreenDiamond im not expert in this but if you read here >> https://soybase.org/correspondence/ you'll see that the desinence "Wm82.a2.v1" means "Williams 82 assembly version 2 annotation version 1". The thing is that across the genome annotation versions your ID's may change. This results in not finding what you are searching for simply because your are searching with an outdated gene/transcript or whatever ID or viceversa you could be searching for something old with a new generated one. THats why on this page you have the correspondence table which allows you to recover what you are searching for across the genome annotation versions. Again citing the example in the link i gave you you have: For example, for Williams 82 assembly version 2 annotation version 1
Locus: Glyma.01g000100
Locus ID: Glyma.01g000100.Wm82.a2.v1
Transcript: Glyma.01g000100.1
Transcript ID: Glyma.01g000100.1.Wm82.a2.v1
From what ive seen they computationally transfer annotations. This is why of course the experimentally validated genes are a minimal part and generally come from Arabidopsis. Anyway if i go on Soybase https://www.soybase.org/ and i search in the SoyBase Toolbox the gene locus ID you are providing i find it >> https://www.soybase.org/sbt/search/search_results.php?category=FeatureName&version=Glyma2.0&search_term=Glyma.18g092200