Notes from email correspondence with Haobo Jiang regarding his 2018 paper "Building a platform for predicting functions of serine protease-related proteins in Drosophila melanogaster and other insects" (FBrf0240739), particularly with respect to reconciling classifications of S1A family peptidases in that paper with previous publications (Ross et al 2003 FBrf0155784), Shah et al 2008 (FBrf0202030), Veillard et al 2016 (FBrf0230906), the MEROPS database (https://www.ebi.ac.uk/merops/index.shtml,accessed Feb 2019) and InterPro & Gene Ontology annotations within FlyBase (FB2018_06).
1. The following genes were annotated as SPH in Ross et al 2003 (and some of the other papers), but the sequences available at the time were incomplete and lacked the catalytic residues; they are correctly classed as SP in Cao & Jiang 2018: CG18636, CG14227, CG17477, CG30028, CG18030, CG1505, CG18125, CG11664
2. These genes wrongly have a SPH annotation in MEROPS; they are correctly classed as SP in Cao & Jiang 2018: CG8172, CG30187, CG30082
3. The following genes are wrongly classed as SP in Shah et al. 2008 and/or MEROPS; they are correctly classed as SPH in Cao & Jiang 2018: CG14990, CG12388, CG8738, CG17242, CG3505, CG5390, CG8586, CG4998, CG18477, CG13318, CG18420, CG4793, CG31326, CG31266, CG31267, CG4653, CG3088, CG31205, CG31780, CG33127, CG33225, CG40160, CG15002
4. These SP/SPH genes appear to be missing in MEROPS: CG17234 (SP), CG32376 (SP), CG34295 (SPH)
5. These genes are classed as SP/SPH in other sources, but are not included in Cao & Jiang 2018 for the given reasons:
- CG15046 (SP in Shah et al 2008, an SPH in MEROPS, has InterPro IPR009003) - this is an SPH and was omitted from our paper in error
- CG8464 (SP in Shah et al 2008 and in MEROPS; has InterPro IPR009003 & IPR001940)- this is an S1C serine protease
- CG10882 (SP in Shah et al 2008) - this is not an SP, presumably an error
- CG31309/CG33202 (SP in Shah et al 2008)  - this is not an SP, presumably an error
- CG34043 (has InterPro IPR009003) - this sequence has only low similarity to S1A and its SP-like domain (195 aa) is shorter than a typical S1A SPH (~230 aa).
- CG3803 (has InterPro IPR009003) - A protein with eight transmembrane regions;  might be a membrane SP(H) but I don’t think so. It has nothing to do with S1A.
- CG3589 (has InterPro IPR009003 & is a SP in MEROPS) This is an S1C (not S1A) serine protease.
- CG3373 (Hemomucin, SPH210 - SPH in Ross et al 2003 & Shah et al 2008) - Not even a SPH, the region similar to SPH is so short.
- CG14218 (SPH211 - SPH in Ross et al 2003) - Not even a SPH, the region similar to SPH is so short.