CupOFMoCA: Coupled Objective-Guided Discrete Flows for Molecular Conjugate Assembly
Organizations: Department of Computer and Information Science, University of Pennsylvania · College of Arts and Sciences, University of Pennsylvania · Department of Bioengineering, University of Pennsylvania
Abstract
Molecular conjugates, including PROTACs and peptide-drug conjugates (PDCs), derive their function from the joint behavior of multiple coupled components, yet most generative approaches design these components independently and combine them only after generation. Such staged pipelines ignore cross-component dependencies and often produce conjugates that are chemically invalid, fall outside empirical conjugate distributions, or lose function upon assembly. In this work, we introduce Coupled Objective-Guided Discrete Flows for Molecular Conjugate Assembly (CupOFMoCA), a discrete generative method that enforces cross-component chemical constraints during conjugate generation. CupOFMoCA restricts guided updates to chemically feasible conjugate states and reweights local transitions using molecular design objectives. Across three PDC systems, coupled generation achieved 100% conjugation validity among valid peptides and increased valid yield over matched post-hoc filtering, with comparable predicted binding scores across the full valid sets. In three-seed evaluations of three PROTAC systems, CupOFMoCA achieved the highest assembly validity on two systems and generated more assembly-valid molecules per second than the evaluated baselines. We further evaluated the generated PROTACs with post-hoc DegradeMaster scoring and explicit-solvent molecular dynamics, supporting their predicted activity and conformational properties. Together, these results establish CupOFMoCA as an efficient generator of molecular conjugates, supporting therapeutic development via coordinated control of assembly and molecular properties.
Figures & tables
| ERBB2 | GNRHR | NRP1 | ||||
| Method | Conj. Val. | Binding | Conj. Val. | Binding | Conj. Val. | Binding |
| moPPIt | 7.1% | 7.59 (0.07) | 8.1% | 6.78 (0.46) | 65.0% | 7.94 (0.06) |
| BoltzGen | ||||||
| Exp | 51.0% | 7.81 (0.05) | 16.0% | 6.63 (1.05) | - | - |
| AF | 21.2% | 7.82 (0.11) | 3.0% | 7.44 (0.45) | 13.7% | 7.05 (1.81) |
| AReUReDi | ||||||
| ERBB2 | GNRHR | NRP1 | ||||
| Method | Conj. Val. | Binding | Conj. Val. | Binding | Conj. Val. | Binding |
| PepTune | 20.8% | 6.75 (0.27) | 5.6% | 4.55 | 13.7% | 7.05 (1.81) |
| AReUReDi | ||||||
| Staged | 30.1% | 7.73 (0.11) | 26.8% | 7.69 (0.75) | 23.0% | 8.62 (0.50) |
| Guided | 100.0% | 8.02 (0.19) | 100.0% | 8.56 (0.35) | 100.0% | 9.16 (0.36) |
| CupOFMoCA | ||||||
| EGFR-VHL | BRD9-VHL | CDK6-CRBN | ||||
| Method | Std. Val. | Asm. Val. | Std. Val. | Asm. Val. | Std. Val. | Asm. Val. |
| LinkerNet | 98.0 | 46.0 | 100.0 | 16.0 | 100.0 | 96.0 |
| DiffPROTACs | 98.0 | 80.0 | 100.0 | 0.0 | 86.0 | 74.0 |
| DiffLinker | 86.0 | 0.0 | 74.0 | 0.0 | 92.0 | 0.0 |
| Ours | 95.3 2.1 | 85.0 3.6 | 94.3 2.1 | 82.3 4.0 | 94.7 1.5 | 83.0 4.2 |
| DegradeMaster predicted degradation probability | ||||||
| EGFR-VHL | BRD9-VHL | CDK6-CRBN | |||||||
| Method | Div | Sim | Div | Sim | Div | Sim | |||
| LinkerNet | 0.895 | 0.816 | 78.3% | 0.829 | 0.825 | 75.0% | 0.833 | 0.776 | 45.8% |
| DiffPROTACs | 0.763 | 0.779 | 35.0% | - | - | - | 0.735 | 0.694 | 2.7% |
| Ours | 0.772 | 0.630 | 0.0% | 0.746 | 0.777 | 26.0% | 0.716 | 0.545 | 0.0% |
Appendix figures & tables34 assets
Supplementary material from the paper’s appendix.
Appendix
| Length | Configuration | ERBB2 | GNRHR | NRP1 |
| 80 | Unconditional | 22.8 / 6.12 | 22.8 / 4.48 | 6.1 / 5.20 |
| Staged | 22.7 / 7.70 | 18.3 / 8.09 | 23.7 / 8.68 | |
| Gate-only | 100.0 / 7.33 | 100.0 / 6.59 | 100.0 / 7.88 | |
| Guided | 100.0 / 7.69 | 100.0 / 7.55 | 100.0 / 8.58 | |
| 100 | Unconditional | 24.5 / 6.22 | 24.5 / 4.93 | 5.0 / 5.47 |
| Staged | 36.7 / 7.92 | 34.7 / 8.15 | 28.1 / 8.65 |
| Group | Mean NLL (SEM) | NLL | 95% CI | |
| PDCdb, all | 393 | 1.5052 (0.0090) | [ , 0.0017] | |
| PDCdb, CPP-like | 149 | 1.4630 (0.0138) | [ , ] | |
| PDCdb, non-CPP | 244 | 1.5349 (0.0118) | 0.0124 | [ , 0.0364] |
| Pretraining validation | 1,414 | 1.5224 (0.0039) | reference | |
| Method | Best mean affinity | Best valid yield/GPU-hour |
| AReUReDi-Staged | 3 | 1 |
| CupOFMoCA-Staged | 8 | 3 |
| AReUReDi-Guided | 1 | 0 |
| CupOFMoCA-Guided | 0 | 8 |
| ERBB2 | GNRHR | NRP1 | ||||
| Method | Conj. Val. | Binding | Conj. Val. | Binding | Conj. Val. | Binding |
| PepTune | 17.9% | 6.82 (0.50) | 17.2% | 5.85 (1.25) | 11.3% | 7.25(1.11) |
| AReUReDi | ||||||
| Staged | 40.9% | 7.94 (0.16) | 39.1% | 9.15 (0.35) | 18.8% | 8.75 (0.64) |
| Guided | 100.0% | 7.97 (0.20) | 100.0% | 8.86 (0.32) | 100.0% | 9.23 (0.24) |
| CupOFMoCA | ||||||
| ERBB2 | GNRHR | NRP1 | ||||
| Method | Conj. Val. | Binding | Conj. Val. | Binding | Conj. Val. | Binding |
| moPPIt | 7.1% | 7.59 (0.07) | 8.1% | 6.78 (0.46) | 65.0% | 7.73 (0.34) |
| BoltzGen | ||||||
| Exp | 51.0% | 7.64 (0.22) | 16.0% | 6.63 (1.05) | - | - |
| AF | 21.2% | 7.80 (0.13) | 3.0% | 7.44 (0.45) | 13.7% | 7.05 (1.81) |
| AReUReDi | ||||||
| ERBB2 | GNRHR | NRP1 | ||||
| Method | Conj. Val. | Binding | Conj. Val. | Binding | Conj. Val. | Binding |
| PepTune | 20.8% | 6.75 (0.27) | 5.6% | 4.55 | 13.7% | 7.05 (1.81) |
| AReUReDi | ||||||
| Staged | 30.1% | 7.65 (0.16) | 26.8% | 7.54 (0.86) | 23.0% | 8.62 (0.50) |
| Guided | 100.0% | 7.68 (0.25) | 100.0% | 7.64 (0.68) | 100.0% | 8.59 (0.50) |
| CupOFMoCA | ||||||
| ERBB2 | GNRHR | NRP1 | ||||
| Method | Conj. Val. | Binding | Conj. Val. | Binding | Conj. Val. | Binding |
| PepTune | 17.9% | 6.82 (0.50) | 17.2% | 5.85 (1.25) | 11.3% | 7.25 (1.11) |
| AReUReDi | ||||||
| Staged | 40.9% | 7.79 (0.21) | 39.1% | 8.57 (0.93) | 18.8% | 8.75 (0.64) |
| Guided | 100.0% | 7.69 (0.21) | 100.0% | 7.95 (0.63) | 100.0% | 8.66 (0.66) |
| CupOFMoCA | ||||||
| Target | Linker-payload (reaction) | Theoretical | Pharma | Generated |
| ( ) | ( ) | |||
| ERBB2 | SSCpt, ester/carbonate (amide) | 537.10 / 553.10 | 538.8 (1) | (92) |
| GNRHR | AoaDau (amide) | 582.18 | (4) | (94) |
| NRP1 | DSPASN38, pos10/pos20 (disulfide) | 478.12 | (4) | (78) |
| Target | Pharma reference | Generated (80 SMILES tokens) | Direction of offset | |
| peptide MW (Da) | seq. length | peptide MW (Da) | ||
| ERBB2 | 1852 ( ) | 18 | 1531 (1205-1986) | generated lighter than pharma |
| GNRHR | 1559 (1072-2300) | 13.2 (8-21) | 1544 (1218-1907) | matched (within Da) |
| NRP1 | 744 (577-974) | 6.8 (5-10) | 1396 (1123-1671) | generated heavier than pharma |
| BoltzGen | AReUReDi | CupOFMoCA | |||||||||
| Property | moPPIt | Exp | AF | Staged (AA) | Guided (AA) | Staged (UAA) | Guided (UAA) | Staged (AA) | Guided (AA) | Staged (UAA) | Guided (UAA) |
| Conj. Validity | 7.1% | 51.0% | 21.2% | 17.0% | 100.0% | 3.1% | 100.0% | 15.0% | 100.0% | 5.1% | 100.0% |
| Binding affinity | 7.59 (0.07) | 7.64 (0.22) | 7.80 (0.13) | 7.65 (0.10) | 7.57 (0.16) | 7.68 (0.08) | 7.72 (0.08) | 7.56 (0.19) | 7.56 (0.14) | 7.77 (0.05) | 7.71 (0.09) |
| MW (Da) | 1797.19 (94.33) | 1862.70 (262.50) | 1652.95 (339.07) | 1958.14 (104.89) | 1987.75 (110.63) | 2460.77 (288.16) | 2558.94 (454.66) | 2062.10 (90.96) | 2037.29 (98.27) | 2238.06 (380.48) | 2465.91 (404.95) |
| LogP | 2.23 (1.93) | (2.37) | (2.89) | 0.93 (1.94) | 0.53 (2.54) | 8.93 (2.53) | 8.96 (5.73) | 0.89 (2.23) | 0.52 (1.99) | 4.59 (3.27) | 6.65 (5.25) |
| TPSA (Å 2 ) | 556.56 (30.58) | 663.67 (120.94) | 571.81 (171.79) | 648.83 (78.14) | 671.25 (83.64) | 668.58 (64.45) | 723.24 (97.92) | 708.49 (84.67) | 698.27 (78.00) | 670.06 (82.08) | 724.58 (111.19) |
| BoltzGen | AReUReDi | CupOFMoCA | |||||||||
| Property | moPPIt | Exp | AF | Staged (AA) | Guided (AA) | Staged (UAA) | Guided (UAA) | Staged (AA) | Guided (AA) | Staged (UAA) | Guided (UAA) |
| Conj. Validity | 8.1% | 16.0% | 3.0% | 17.6% | 100.0% | 5.1% | 100.0% | 15.0% | 100.0% | 4.0% | 100.0% |
| Binding affinity | 6.78 (0.46) | 6.63 (1.05) | 7.44 (0.45) | 6.58 (0.42) | 6.48 (0.49) | 7.68 (0.33) | 7.39 (0.47) | 6.30 (0.35) | 6.40 (0.50) | 6.98 (0.84) | 7.57 (0.45) |
| MW (Da) | 1807.91 (122.97) | 1914.95 (267.16) | 1629.48 (178.58) | 2057.01 (112.57) | 2044.31 (106.20) | 2431.32 (446.81) | 2506.25 (395.02) | 2081.44 (99.03) | 2074.30 (110.34) | 2852.63 (370.48) | 2650.75 (445.82) |
| LogP | (3.66) | (3.00) | 0.36 (1.88) | (1.74) | (1.40) | 2.62 (3.27) | 4.77 (6.72) | (1.42) | (1.79) | 8.73 (11.32) | 7.00 (8.38) |
| TPSA (Å 2 ) | 630.78 (69.49) | 712.12 (138.16) | 582.29 (78.16) | 794.03 (91.01) | 782.39 (73.53) | 835.99 (164.41) | 806.31 (98.75) | 787.20 (74.23) | 795.54 (86.98) | 836.49 (137.88) | 802.97 (104.47) |
| AReUReDi | CupOFMoCA | |||||||||
| Property | moPPIt | BoltzGen AF | Staged (AA) | Guided (AA) | Staged (UAA) | Guided (UAA) | Staged (AA) | Guided (AA) | Staged (UAA) | Guided (UAA) |
| Conj. Validity | 65.0% | 13.7% | 19.0% | 98.0% | 37.4% | 100.0% | 25.0% | 100.0% | 19.4% | 100.0% |
| Binding affinity | 7.73 (0.34) | 7.05 (1.81) | 8.23 (0.38) | 8.24 (0.37) | 7.92 (0.32) | 7.94 (0.27) | 8.16 (0.32) | 8.28 (0.42) | 8.02 (0.40) | 7.95 (0.22) |
| MW (Da) | 1754.71 (96.21) | 1680.54 (141.90) | 1838.20 (115.56) | 1842.31 (82.55) | 2431.86 (271.82) | 2325.85 (252.47) | 1854.46 (94.83) | 1859.13 (87.62) | 2269.39 (317.96) | 2244.83 (296.12) |
| LogP | 4.11 (2.09) | 2.22 (5.38) | 2.87 (2.07) | 3.00 (2.07) | 10.74 (4.53) | 8.75 (3.94) | 3.73 (2.07) | 3.40 (2.16) | 10.37 (4.20) | 8.23 (4.44) |
| TPSA (Å 2 ) | 525.45 (55.18) | 518.17 (130.08) | 566.37 (70.85) | 566.61 (55.68) | 626.38 (64.34) | 625.54 (78.18) | 556.44 (65.69) | 566.99 (72.48) | 569.27 (61.28) | 596.79 (61.02) |
| Property | PepTune | AReUReDi Staged | AReUReDi Guided | CupOFMoCA Staged | CupOFMoCA Guided |
| Conj. Validity | 20.8% | 30.1% | 100.0% | 22.7% | 100.0% |
| Binding affinity | 6.75 (0.27) | 7.65 (0.16) | 7.68 (0.25) | 7.70 (0.17) | 7.69 (0.22) |
| MW (Da) | 1875.21 (110.72) | 2117.46 (121.51) | 2062.85 (150.76) | 2159.62 (125.87) | 2069.22 (156.24) |
| LogP | (5.03) | (3.55) | (5.32) | (4.08) | (5.04) |
| TPSA (Å 2 ) | 665.82 (97.57) | 807.02 (101.27) | 750.90 (142.53) | 779.71 (92.59) | 750.40 (131.44) |
| HBD | 18.60 (4.03) | 25.64 (3.64) | 25.01 (5.26) | 25.05 (3.36) | 24.35 (4.74) |
| Property | PepTune | AReUReDi Staged | AReUReDi Guided | CupOFMoCA Staged | CupOFMoCA Guided |
| Conj. Validity | 5.6% | 26.8% | 100.0% | 18.3% | 100.0% |
| Binding affinity | 4.55 | 7.54 (0.86) | 7.64 (0.68) | 8.09 (1.17) | 7.55 (0.68) |
| MW (Da) | 1857.06 (0.00) | 2161.09 (140.02) | 2136.08 (132.55) | 2100.42 (90.95) | 2128.04 (142.41) |
| LogP | (3.92) | (4.20) | (4.57) | (4.02) | |
| TPSA (Å 2 ) | 777.09 | 814.26 (87.94) | 821.02 (98.05) | 791.03 (105.28) | 795.98 (105.34) |
| HBD | 27.00 | 30.45 (3.73) | 29.63 (3.96) | 27.76 (3.70) | 28.76 (4.21) |
| Property | PepTune | AReUReDi Staged | AReUReDi Guided | CupOFMoCA Staged | CupOFMoCA Guided |
| Conj. Validity | 13.7% | 23.0% | 100.0% | 23.7% | 100.0% |
| Binding affinity | 7.05 (1.81) | 8.62 (0.50) | 8.59 (0.50) | 8.68 (0.37) | 8.58 (0.64) |
| MW (Da) | 1680.54 (141.90) | 1928.14 (132.97) | 1907.99 (115.10) | 1925.42 (95.30) | 1875.63 (124.12) |
| LogP | 2.22 (5.38) | 3.93 (3.67) | 4.98 (3.32) | 5.09 (3.89) | 4.88 (3.55) |
| TPSA (Å 2 ) | 518.17 (130.08) | 576.12 (85.53) | 528.84 (72.29) | 543.50 (68.90) | 528.09 (81.09) |
| HBD | 14.71 (5.09) | 19.45 (2.86) | 17.40 (2.96) | 17.61 (4.05) | 17.82 (3.58) |
| Property | PepTune | AReUReDi Staged | AReUReDi Guided | CupOFMoCA Staged | CupOFMoCA Guided |
| Conj. Validity | 17.9% | 40.9% | 100.0% | 36.7% | 100.0% |
| Binding affinity | 6.82 (0.50) | 7.79 (0.21) | 7.69 (0.21) | 7.92 (0.25) | 7.86 (0.27) |
| MW (Da) | 2135.35 (135.67) | 2458.65 (158.09) | 2422.54 (176.41) | 2513.95 (147.00) | 2436.57 (171.09) |
| LogP | 0.92 (3.59) | (4.20) | (4.88) | (5.23) | (5.44) |
| TPSA (Å 2 ) | 674.01 (54.77) | 905.03 (103.73) | 887.69 (136.12) | 913.99 (110.20) | 875.13 (131.86) |
| HBD | 20.80 (3.56) | 28.56 (4.04) | 29.03 (4.76) | 28.17 (5.85) | 28.34 (5.09) |
| Property | PepTune | AReUReDi Staged | AReUReDi Guided | CupOFMoCA Staged | CupOFMoCA Guided |
| Conj. Validity | 17.2% | 39.1% | 100.0% | 34.7% | 100.0% |
| Binding affinity | 5.85 (1.25) | 8.57 (0.93) | 7.95 (0.63) | 8.15 (0.74) | 8.07 (0.74) |
| MW (Da) | 2180.29 (135.67) | 2534.01 (131.60) | 2420.67 (220.22) | 2544.61 (111.73) | 2536.91 (143.44) |
| LogP | (3.59) | (3.84) | (4.31) | (4.25) | (4.13) |
| TPSA (Å 2 ) | 750.86 (54.77) | 981.23 (93.21) | 953.53 (119.62) | 952.33 (119.41) | 962.22 (105.10) |
| HBD | 24.80 (3.56) | 33.76 (4.20) | 32.85 (4.86) | 32.76 (4.34) | 32.96 (4.70) |
| Property | PepTune | AReUReDi Staged | AReUReDi Guided | CupOFMoCA Staged | CupOFMoCA Guided |
| Conj. Validity | 11.3% | 18.8% | 100.0% | 28.1% | 100.0% |
| Binding affinity | 7.25(1.11) | 8.75 (0.64) | 8.66 (0.66) | 8.65 (0.53) | 8.67 (0.47) |
| MW (Da) | 2001.96(143.32) | 2270.30 (179.81) | 2327.55 (147.16) | 2243.58 (171.81) | 2285.06 (149.25) |
| LogP | 0.04 (2.32) | 4.29 (4.16) | 3.20 (5.86) | 4.36 (4.79) | 2.85 (5.81) |
| 683.09 (80.04) | 698.83 (110.29) | 731.19 (147.33) | 684.12 (140.35) | 711.35 (153.47) | |
| HBD | 19.83 (1.72) | 22.50 (4.10) | 24.25 (5.10) | 22.26 (4.74) | 23.29 (5.64) |
| Property | GNRHR-Dau Coupled | GNRHR-Dau Var. Linker | ERBB2-Cpt Coupled | ERBB2-Cpt Var. Linker |
| Conj. validity | 100% | 100% | 100% | 100% |
| Binding affinity | 7.55 (0.68) | 7.67 (0.96) | 7.69 (0.22) | 7.73 (0.23) |
| MW (Da) | 2128 (142) | 2136 (178) | 2069 (156) | 2090 (149) |
| LogP | (4.02) | (3.92) | (5.04) | (4.98) |
| TPSA (Å 2 ) | 796.0 (105.3) | 796.6 (102.1) | 750.4 (131.4) | 760.9 (137.8) |
| HBD | 28.8 (4.2) | 28.9 (4.2) | 24.4 (4.7) | 24.2 (4.6) |
| Receptor | Variant | Length | Validity | Uniqueness | Diversity | SNN |
| ERBB2 | Coupled | 80 | 0.919 | 1.000 | 0.844 | 0.187 |
| 100 | 0.949 | 1.000 | 0.835 | 0.194 | ||
| Var. linker | 80 | 0.919 | 1.000 | 0.842 | 0.188 | |
| 100 | 0.960 | 1.000 | 0.835 | 0.194 | ||
| GNRHR | Coupled | 80 | 0.939 | 1.000 | 0.843 | 0.180 |
| 100 | 0.960 | 1.000 | 0.832 | 0.187 |
| Metric | Peptide | Conjugate | ||
| Pharma | Ours | Pharma | Ours | |
| Conf. | 0.6653 | 0.6878 | 0.6415 | 0.6971 |
| ipTM | 0.7217 | 0.7571 | 0.6281 | 0.8043 |
| pLDDT | 0.6511 | 0.6705 | 0.6448 | 0.6703 |
| ipLDDT | 0.4344 | 0.4976 | 0.4021 | 0.5023 |
| Metric | Peptide | Conjugate | ||
| Pharma | Ours | Pharma | Ours | |
| Conf. | 0.8037 | 0.7764 | 0.7111 | 0.7251 |
| ipTM | 0.9217 | 0.8826 | 0.8581 | 0.8695 |
| pLDDT | 0.7742 | 0.7498 | 0.6744 | 0.6890 |
| ipLDDT | 0.6522 | 0.5827 | 0.4639 | 0.4961 |
| Metric | Peptide | Conjugate | ||
| Pharma | Ours | Pharma | Ours | |
| Conf. | 0.7314 | 0.6545 | 0.6560 | 0.6356 |
| ipTM | 0.6673 | 0.3914 | 0.4359 | 0.3486 |
| pLDDT | 0.7474 | 0.7203 | 0.7111 | 0.7074 |
| ipLDDT | 0.6563 | 0.4421 | 0.4547 | 0.4036 |
| Target | Property | LinkerNet | DiffPROTACs | Ours |
| EGFR-VHL | SA score | 1.89 0.46 | 2.95 0.91 | 2.76 0.57 |
| QED | 0.487 0.072 | 0.424 0.133 | 0.51 0.13 | |
| in range | 87.0% | 75.0% | 70.3% | |
| MW in range | 100.0% | 80.0% | 71.4% | |
| RotB in range | 100.0% | 67.5% | 84.5% | |
| Uniqueness | 65.2% | 97.5% | 96.2% |
| Linker | Role | 3D PSA (Å 2 ) | Penalty | |
| C1 | -propylsuccinyl | Generated | 140 | 2.2 |
| C2 | allyl succinyl | Generated | 141 | 1.9 |
| C3 | gem-dimethylsuccinate | Generated | 178 | 2.7 |
| C4 | butylmalonate | Generated | 184 | 3.3 |
| MS39 | decanedioyl | Positive | 178 | 3.7 |
| PEG4 | PEG4 | Negative | 193 | 4.1 |
| Parameter | Value |
| Hardware | 4 A6000 NVIDIA RTX 6000 Ada GPU (48 GB VRAM) |
| Training epochs | 5 |
| Optimizer | Adam |
| Learning rate | |
| Learning rate schedule | Cosine decay with 10% warmup |
| Minimum learning rate |
| Component | Configuration |
| Input | Unpooled token-level embeddings from ESM-2 650M |
| Convolutional layers | 3 initial CNN kernel layers (dimension 384) |
| Cross-attention layers | 4 layers (dimension 2048) |
| Prediction head | Shared prediction head (dimension 1024) |
| Parameter | Value |
| Training epochs | 50 |
| Learning rate | |
| Dropout rate | 0.15 |
| CNN layers | 3 initial layers (dimension 384) |
| Cross-attention layers | 4 layers (dimension 2048) |
| Prediction head dimension | 1024 |
| Method | SMARTS Pattern | Purpose |
| has_aoa_amide_conjugation() | [NX3]CX3[CH2][OX2][NX2]=[C] |
Molecular Generation
Molecular generation develops computational methods to create chemical structures and conformations for discovery, optimization, and structure elucidation.
See all 252 papers