Home Research Services
Online Services
STOPred
StoPred result

example1

model_trainingModelCurrent release (cutoff 2026-01-01)scheduleGenerated2026-10-28 03:23arrow_backBack to StoPred
Predicted stoichiometry
A6
Ranking score 0.61
Next: A12 (0.16), A2 (0.13)
6AT1270-1
Copies12345-67+
Top PDB template: 1KY9 assembly 1 · A×6 · see Templates
Homo-oligomer prediction
Score 0.892

Probability of any copy number other than 1, from the monomer-aware model. Probabilities for each copy number are under Assembly-state details.

Drag the red threshold to see the trade-off
0.500
0.10.250.50.750.90.99
Recall
81%
Precisioninfo_outlinePrecision depends on how common homo-oligomers are among the sequences tested. The benchmark has 742 homo-oligomers among 1,231 chains (60%); precision is lower when homo-oligomers are rarer among the sequences you test.
83%
False-positive rate
25%

Estimated on the evaluation dataset.

Default 0.5 (release 2026-01-01).

Stoichiometry ranking

RankStoichiometryRanking scoreCopies per chain
1A6
0.6139
A T1270-1: 6
2A12
0.1629
A T1270-1: 12
3A2
0.1253
A T1270-1: 2
4A3
0.0545
A T1270-1: 3
5A8
0.0144
A T1270-1: 8
6A14
0.0077
A T1270-1: 14
7A4
0.0067
A T1270-1: 4
8A7
0.0062
A T1270-1: 7
9A24
0.0018
A T1270-1: 24
10A10
0.0014
A T1270-1: 10

Per-chain copy numbers

Probability of each copy number for every chain, from the chain-level head. "other" is a copy number outside the trained classes.

AT1270-1437 aa
6
0.614
12
0.163
2
0.125
3
0.055
8
0.014
14
0.008

Assembly-state details

Copy-number probabilities from the monomer-aware model. "1" means monomer.

6
0.429
12
0.258
2
0.138
1
0.108
3
0.037
8
0.010
4
0.008
24
0.006
7
0.002
10
0.001

Templates

Sequence search (MMseqs) of the submitted chains against PDB biological assemblies (template database StoPred_template_db_20260505, PDB snapshot 2026-05-05). Query chains that a template deposits as one chain (merge) or as several chains (split) are matched as one unit. 108 templates found; the top 50 are listed. Click a row for the structure and the alignments.

Limited to PDB entries released before 2026-01-01 (0 later templates left out).

#TemplateScoreIdentityE-valueQuery positionTemplate positionTemplate copy numbers
11KY9 asm 1
75A 35%4e-60
A
A×6
23OU0 asm 1
75A 35%4e-60
A
A×24
33MH5 asm 1
74A 36%1e-60
A
A×6
43MH6 asm 1
74A 36%1e-60
A
A×24
53MH4 asm 1
73A 35%4e-60
A
A×6
64A9G asm 1
70A 33%1e-58
A
A×24
74A8C asm 1
70A 33%1e-58
A
A×12
86Z05 asm 1
66A 33%3e-55
A
A×12
97XS2 asm 1
66A 33%3e-54
A
A×1
103PV2 asm 1
64A 33%9e-52
A
A×12
113PV5 asm 1
63A 33%6e-51
A
A×12
123PV4 asm 1
52A 37%3e-47
A
A×3
138K2Y asm 1
49A 38%6e-51
A
A×3
144YO1 asm 1
46A 39%1e-46
A
A×3
153NUM asm 1
40A 34%9e-36
A
A×3
165T69 asm 1
40A 33%2e-40
A
A×6
177W4S asm 1
38A 38%6e-37
A
A×3
187W23 asm 1
38A 38%6e-37
A
A×3
196JJK asm 1
38A 35%4e-60
A
A×12
206JJL asm 1
38A 35%4e-60
A
A×12
212ZLE asm 1
38A 36%1e-60
A
A×12
227W25 asm 1
38A 37%4e-37
A
A×3
237W4V asm 1
38A 37%4e-37
A
A×3
247W21 asm 1
38A 37%6e-37
A
A×3
256IEO asm 1
38A 37%6e-37
A
A×1
267W4R asm 1
38A 37%6e-37
A
A×3
277VYZ asm 1
38A 37%6e-37
A
A×1
287W4T asm 1
38A 37%6e-37
A
A×3
294A8D asm 1
37A 35%4e-60
A
A×12
307VZ0 asm 1
37A 37%1e-36
A
A×1
317W4U asm 1
37A 37%1e-36
A
A×3
325ZVJ asm 1
37A 37%1e-36
A
A×1
337W4W asm 1
37A 37%3e-36
A
A×3
347W22 asm 1
37A 37%3e-36
A
A×3
357W24 asm 1
37A 37%6e-36
A
A×3
363OTP asm 1
36A 35%4e-60
A
A×12
376JJO asm 1
36A 35%4e-60
A
A×12
384A8A asm 1
35A 33%1e-58
A
A×12
394A8B asm 1
35A 33%1e-58
A
A×12
401L1J asm 1
34A 41%6e-39
A
A×1
413STI asm 1
33A 43%3e-39
A
A×3
424IC6 asm 1
29A 31%5e-32
A
A×6
434RI0 asm 1
26A 34%7e-26
A
A×3
445JD8 asm 1
26A 30%1e-28
A
A×3
455TO1 asm 1
25A 32%6e-25
A
A×3
461Y8T asm 1
24A 32%4e-24
A
A×3
471VCW asm 1
23A 29%3e-27
A
A×3
484RR0 asm 1
23A 29%3e-27
A
A×3
494RQY asm 1
23A 29%3e-27
A
A×3
504RR1 asm 1
23A 29%3e-27
A
A×3

Input sequences

AT1270-1437 aa
1EEKDNTVRALQDSFREVSKKILPSSVEVHATGVIKQSFPVPFFFFDMPEFDSERKSNWAG
61SGVIIGRDSQKKSLFYVVTNSHVVDKATELEVVSYDKKKHKAKLVGKDEKKDIALISFES
121DDATIKVADLGDSDKLEIGDWVMAVGSPFQFSFTVTAGIVSGLQRSANPNLQSRNLFIQT
181DAAINRGNSGGPLVNIKGEVIGINAWIASNSGGNIGLGFAIPVNNIKSTVDFFLKGKKIE
241SAWLGISFYPLKTRDSEVLKSLGVESNDVSAAIIASLYPGSPAVKSGLRAGDIIMKVNGV
301SMSVFQDVTSYISDFYAGEKVNVEILRGNVKKNIEIVLAVRPKDKELSSSKMLPGFVVYP
361LVEDIKAQLNLRNWIKGVVVDYIDKNLASNIKMKSGDVILSVNSKSVSNLREFYDALEVG
421KNTYKILRGNDSFKITF
How to read this page
  • Stoichiometry assumes the submitted sequences form one complex and ranks copy numbers per chain (A2B1 = two copies of chain A, one of chain B). It is always reported.
  • Assembly-state check tests that assumption. For one sequence, a monomer-aware model estimates whether it is a monomer or a homo-oligomer. For 2–6 distinct sequences, a co-assembly model scores whether they occur in one assembly.
  • Scores are model estimates from sequence alone (ESM-C embeddings), not experimental evidence.
Liu, Q., Peng, C., Zheng, W., Zhang, C., & Freddolino, L. (2025). StoPred: Accurate Stoichiometry Prediction for Protein Complexes Using Protein Language Models and Graph Attention. bioRxiv. https://doi.org/10.1101/2025.10.20.683515

FreddolinoLabumich.edu | 1150 W. Medical Center Dr, Ann Arbor, MI 48109-0600