Initial Falcon pass, 2026-05-12. HRAS is a fully pending review, so I kept the
first curation pass focused on the most defensible core biology rather than
attempting all 258 annotations at once.
Core function: HRAS is a Ras-family small GTPase. UniProt summarizes that Ras
proteins bind GDP/GTP and have intrinsic GTPase activity [file:human/HRAS/HRAS-uniprot.txt
"Ras proteins bind GDP/GTP and possess intrinsic GTPase activity
(PubMed:12740440, PubMed:14500341, PubMed:9020151)."]. Falcon likewise frames
HRAS as a GDP/GTP molecular switch [file:human/HRAS/HRAS-deep-research-falcon.md
"RAS proteins (including HRAS) are GTP hydrolases that function as
binary molecular switches: GDP-bound is “OFF,” GTP-bound is “ON,” and
the ON state supports binding to downstream effector proteins."].
Core processes: Ras protein signal transduction and positive regulation of the
MAPK cascade are central. Falcon specifically recommends these as core HRAS
terms but warns against defaulting distal outcomes to core annotations
[file:human/HRAS/HRAS-deep-research-falcon.md "“Ras protein signal
transduction” and “positive regulation of MAPK cascade” are generally core
for HRAS, but very specific outcome terms (e.g., “cell cycle progression” as a
default) should be restricted to contexts with strong, direct, wild-type
evidence."].
Localization: HRAS membrane association is driven by CAAX processing and
palmitoylation, so plasma membrane and Golgi membrane are the safest core
locations in this first pass [file:human/HRAS/HRAS-deep-research-falcon.md
"HRAS contains a C-terminal CAAX motif that is post-translationally
processed by a canonical sequence: farnesylation → AAX proteolysis (RCE1)
→ carboxymethylation (ICMT)."].
Non-core and over-annotation cautions: proliferation, senescence, cell-cycle,
motility, transcriptional, and developmental terms are often downstream or
mutant/oncogenic-context phenotypes. PMID:9054499 is explicit that its senescence
evidence used oncogenic Ras PMID:9054499. PMID:23027131 likewise uses a malignant transformation model
PMID:23027131.
PLCE1 effector branch: PMID:11022048 supports direct GTP-dependent HRAS binding
and activation of PLCepsilon, so I kept GO:0160185 as a non-core effector branch
rather than core function PMID:11022048.
Left for later: most generic protein binding annotations remain PENDING. The
right follow-up is to replace individual generic protein-binding rows with
specific effector-binding terms only where there is direct, nucleotide-state
dependent binding evidence.