- Fold
- Twelve predicted transmembrane helices, an MFS-fold architecture; multi-pass lysosomal-membrane transporter; non-enzyme.
- Substrate Cavity
- A residue-level substrate-cavity hypothesis from AstraBIND (confidence 0.79, HIGH), unclassified — a starting point for docking and cavity mutagenesis, not a validated ligand.
- Structural Anchors
- Predicted N-glycosylation at N371 and N376 (UniProt), on the luminal face of the transporter.
- Flexible Regions
- Elevated predicted disorder at the cytosolic N-terminus (1–21), a cytosolic loop (246–256) and the luminal loop (371–396) — natural truncation boundaries for construct design.
- Clean Signal
- One aggregation-prone segment flagged (AstraUNFOLD amyloid max 0.556) — worth screening in construct design.
Model-reported confidence for the headline calls (amber = the load-bearing prediction the rest of the profile builds on). These are model-estimated probabilities that rank and gate each call — not calibrated rates of experimental success.
The Gap
Why This Target Is Still Dark
Most tractable membrane-transporter families have been structurally explored. MFSD8 — the gene better known as CLN7 — has not: an IDG Tbio lysosomal protein of the major facilitator superfamily, with no experimental structure in the PDB and a thin functional literature. It was first identified as a putative lysosomal transporter, and even its activity is still moving — UniProt now annotates an outward-rectifying chloride-channel function in the endolysosome — while the physiological substrate whose loss drives disease remains unidentified. Yet the interest is unambiguous: recessive loss-of-function mutations cause variant late-infantile neuronal ceroid lipofuscinosis (CLN7, a fatal childhood form of Batten disease), and a distinct set of mutations underlies a nonsyndromic macular dystrophy.
That combination — genuine clinical interest, near-zero structural information — is where prediction earns its keep: everything below is computed from the canonical 518-residue sequence with Orbion's Astra suite, with no experimental MFSD8 structure used as input. For a lysosomal orphan with no solved fold, there is nothing to look up.
Architecture & Topology
How the Sequence Is Organised
| Element | Residues | Note |
|---|---|---|
| Transmembrane Helices | 12 predicted | Boundaries: 41–61; 75–95; 106–126; 132–152; 174–194; 212–232; 267–287; 305–325; 338–358; 413–433; 452–472; 483–503. |
The Predicted Pocket
The Predicted Substrate Cavity
For this genuine orphan, AstraBIND has no ligand-bound relative to retrieve from — so these residues are a structure-based cavity prediction, not a retrieval-grounded pocket, and the score reflects cavity geometry rather than a known binding site. Treat them as an exploratory starting point for mutagenesis only; the topology, disorder and modification maps above are the higher-confidence outputs for this target. No validated ligand; not a proven druggable site.
Site: Predicted central translocation cavity
Post-Translational & Structural Features
Specific, Testable Residues
- N-glycosylation at position 371, 376 (UniProt) — on the luminal loop between TM9 and TM10, the expected site for glycan attachment and a handle for trafficking studies.
- Aggregation-prone segment flagged (AstraUNFOLD amyloid max 0.556) — a predicted amyloidogenic stretch to screen out or engineer around when designing expression constructs.
Recommended Experimental Follow-Up
An Orphan Sequence, Turned Into a Ranked Plan
Each prediction is paired with the experiment that would test it and the readout to watch for.
| Prediction | Experiment | Readout |
|---|---|---|
| Substrate-Cavity Residues (TM7, TM10, TM11) | Cavity-lining alanine scan in a lysosomal transport / halide-flux assay | Loss of transport at the predicted site |
| Disordered N-Terminus and Loops (1–21, 246–256, 371–396) | Fusion-partner insertion or loop truncation | Improved expression / thermostability for cryo-EM |
| Predicted MFS Transporter Class | Substrate-uptake / electrophysiology in a defined lysosomal system | Confirm transport function and identify the cargo |
Scope & Limitations
What This Is — and Isn't
- Prediction, not experiment. These are computational hypotheses to prioritise experiments — not a structure or an assay. Nothing here is wet-lab validated.
- The pocket is predicted; the ligand is not named. AstraBIND is retrieval-based; the honest output is a residue-level hypothesis, not a proven druggable site or a validated binder.
- The substrate is genuinely unresolved. MFSD8/CLN7 has been annotated as both a putative MFS transporter and, more recently, an endolysosomal chloride channel; the physiological cargo whose loss causes CLN7 Batten disease is not established. The cavity hypothesis here names residues to test, not a transported molecule.
All predictions were generated with Orbion's Astra suite from the canonical MFSD8 sequence (UniProt Q8NHS3), using AlphaFold-derived structural features. Reported values are model outputs; model internals are out of scope.
References
- [1]UniProt Consortium. UniProtKB entry Q8NHS3 (MFSD8, human). uniprot.org.
- [2]Pharos (Illuminating the Druggable Genome). MFSD8 target record — Tbio. pharos.nih.gov.
- [3]Siintola E et al. The novel neuronal ceroid lipofuscinosis gene MFSD8 encodes a putative lysosomal transporter. (2007). https://doi.org/10.1086/518902
- [4]Kousi M et al. Mutations in CLN7/MFSD8 are a common cause of variant late-infantile neuronal ceroid lipofuscinosis. (2009). https://doi.org/10.1093/brain/awn366
- [5]Roosing S et al. Mutations in MFSD8, encoding a lysosomal membrane protein, are associated with nonsyndromic autosomal recessive macular dystrophy. (2015). https://doi.org/10.1016/j.ophtha.2014.07.040