The defining molecular activity is nucleotide exchange on CDC42. The human fragment tested in PMID:8969170(../../../publications/PMID_8969170.md) contains residues 330–710, encompassing the DH and first PH domains. Its exchange assay measures GDP release and GTPγS loading, with different responses from Sf9-derived and bacterially expressed CDC42. That distinction matters: exchange-mediated activation should not be described as a measurement of GTP hydrolysis. The direct source guide records the broader original-paper read.
The mouse donor paper PMID:12913069(../../../publications/PMID_12913069.md) supplies the localization and cell-shape evidence behind six manual ortholog transfers. Its Results accompanying Figure 4 distinguish cytoplasmic/Golgi pools from ruffle and lamellipodial recruitment. Figures 6–7 and Table 1 connect altered targeting with abnormal actin patterns and morphology. Methods identify mouse Fgd1 and engineered expression constructs. The inability to obtain endogenous coimmunoprecipitation is stated in the Discussion; the tagged-protein interaction evidence should retain that limit. No direct FGD1–actin-binding activity follows from association with cortactin or mAbp1.
Human HPA immunofluorescence independently supports cytosol and plasma-membrane association. The inspected antibody/sample assignments and two sample images are consistent with extranuclear and peripheral staining. The images do not resolve membrane-spanning topology or quantify soluble versus membrane-associated pools. The second antibody’s U2OS images lack the positive location assignment given to A-431 and PODO/TERT256 in the inspected XML.
The actual PAINT tree places human leaf PTN002489341 below both ancestral nodes used by the four imported IBA rows. The ungapped MSA sequence exactly matches the 961-residue canonical protein. This resolves target placement while leaving the biological interpretation separate: the inherited GEF activity is central, whereas filopodial and morphological responses are downstream cellular consequences. Self-WITH is expected experimental grounding.
Reactome explicitly places FGD1 in the established CDC42 GEF set. The broader p75NTR and RhoA/B/C events reuse a generic GEF set that also contains FGD1. Their imported cytosol annotations remain meaningful; their pathway titles do not establish that FGD1 activates RAC1 or every Rho-family GTPase. All four target routes are recorded in the source-evidence table.
ARBA’s actual target-matching branches depend on DH/PH-associated FunFam features and, for one rule, Primates. InterPro maps the DH region to GEF activity and the FYVE region to metal binding. The latter is a structural-domain inference, not a measurement of target metal affinity or proof of phosphoinositide specificity.
The accepted record has no curated natural splice products. The full 961-residue chain, the human assay fragment and the 960-residue mouse donor must remain distinct. Endogenous targeting and autoinhibition of the full human protein remain more precise questions than assigning every engineered phenotype to a natural isoform.
The genuine provider invocation did not produce a report; no substitute provider report was authored. Original-paper and database reads are documented in the source guide. The donor normal is abstract-only even though selected full original sections were read separately. No new publication quotations or GO annotations were added.