Back

Dissecting the Black Box of AlphaFold in Protein-Protein Complex Assembly

Li, S.; Mu, Z.; Yan, C.

2026-04-06 bioinformatics
10.64898/2026.04.03.716280 bioRxiv
Show abstract

AlphaFold achieves unprecedented accuracy in modeling protein-protein complexes, yet the principles governing complex assembly remain unclear. Here, we develop a unified interpretability framework for AlphaFold-Multimer and AlphaFold3 to dissect the mechanisms underlying complex formation. We demonstrate that inter-protein coevolution is not a major determinant of assembly. Instead, complex structures are primarily driven by monomer geometry together with interface-level pattern matching between backbone complementarity and residue identities. By visualizing the iterative propagation of distance constraints during inference, we uncover a hierarchical process in which monomer-level constraints are established prior to cross-chain interactions, directly demonstrating that inter-chain geometry is inferred from monomer geometries rather than being encoded by coevolutionary signals. Application to antigen-antibody complexes further reveals that reduced prediction accuracy arises from the non-canonical and structurally plastic nature of immune interfaces, identifying accurate modeling of interface conformations and recognition of atypical antigen-antibody interaction patterns as key bottlenecks for improving immune complex prediction.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

1
Cell Systems
167 papers in training set
Top 0.1%
27.7%
2
Nature Communications
4913 papers in training set
Top 10%
14.4%
3
Nature Biotechnology
147 papers in training set
Top 2%
4.9%
4
Proceedings of the National Academy of Sciences
2130 papers in training set
Top 14%
4.9%
50% of probability mass above
5
Science
429 papers in training set
Top 6%
4.9%
6
Nature
575 papers in training set
Top 7%
3.7%
7
Nucleic Acids Research
1128 papers in training set
Top 6%
3.6%
8
eLife
5422 papers in training set
Top 36%
2.1%
9
Advanced Science
249 papers in training set
Top 8%
2.1%
10
PLOS Computational Biology
1633 papers in training set
Top 14%
1.9%
11
Nature Machine Intelligence
61 papers in training set
Top 2%
1.9%
12
Nature Methods
336 papers in training set
Top 4%
1.7%
13
Cell Genomics
162 papers in training set
Top 4%
1.3%
14
Cell Reports
1338 papers in training set
Top 27%
1.3%
15
Science Advances
1098 papers in training set
Top 23%
1.2%
16
Nature Cell Biology
99 papers in training set
Top 4%
1.1%
17
Nature Genetics
240 papers in training set
Top 6%
1.1%
18
Cell
370 papers in training set
Top 15%
1.0%
19
Nature Chemical Biology
104 papers in training set
Top 3%
0.9%
20
Nature Microbiology
133 papers in training set
Top 4%
0.7%
21
Nature Computational Science
50 papers in training set
Top 2%
0.7%
22
Frontiers in Immunology
586 papers in training set
Top 8%
0.7%
23
Science Immunology
81 papers in training set
Top 2%
0.7%
24
Immunity
58 papers in training set
Top 4%
0.7%
25
Developmental Cell
168 papers in training set
Top 13%
0.6%
26
JCI Insight
241 papers in training set
Top 8%
0.6%
27
Molecular Biology and Evolution
488 papers in training set
Top 5%
0.6%
28
Genome Biology
555 papers in training set
Top 8%
0.6%