Can Protein-Derived Knowledge Improve Pathology Foundation Models?

Di Zhang; Zhangpeng Gong; Jiashuai Liu; Zhi Zeng; Jiusong Ge; Chunze Yang; Xitong Ling; Kai Yi

cs.CVSep 27, 2026

Can Protein-Derived Knowledge Improve Pathology Foundation Models?

Authors: Di Zhang, Zhangpeng Gong, Jiashuai Liu, Zhi Zeng, Jiusong Ge, Chunze Yang, Xitong Ling, Kai Yi, +5 more

Organizations: School of Computer Science and Technology, Xi’an Jiaotong University · Tsinghua Shenzhen International Graduate School, Tsinghua University · University of Cambridge · Saw Swee Hock School of Public Health, National University of Singapore · Bioinformatics Institute (BII), A*STAR · Department of Oncology, University of Cambridge · Institute of Intelligent Medicine, Peking Union Medical College

Abstract

Molecularly guided pathology foundation models (PFMs) exploit transcriptomic or proteomic information to enrich whole-slide image (WSI) representations, yet effectively leveraging large standalone molecular corpora remains challenging. First, existing molecular foundation models encode protein sequences or single-cell states, not the patient-level bulk expression profiles paired with WSIs. Second, because cross-modal supervision is restricted to paired WSI-omics samples, knowledge from standalone molecular corpora reaches the pathology encoder only indirectly, creating a paired-support bottleneck. To address these challenges, we propose a three-stage framework that decouples proteomic knowledge acquisition from cross-modal transfer, yielding ProSlide, a slide-level hierarchical pathology foundation model. First, to close the modality gap, we pretrain a Proteomic Foundation Encoder (PFE) on 12,695 sample-level bulk protein profiles using virtual profile generation and expression-space multi-view pretraining. Second, we pretrain ProSlide, a patch-region-slide encoder, to predict protein expression from paired WSI-protein samples. Third, to relax the paired-support bottleneck, we introduce Prot2Path, a cross-modal relational distillation objective. For each paired sample, it aligns the similarity distributions of the WSI and its protein profile over a shared, frozen bank of PFE-encoded paired and standalone profiles. We evaluate ProSlide on 12 downstream tasks across breast, lung, and renal cancers. Despite being pretrained with only 2,229 WSIs and 12,695 sample-level protein profiles, ProSlide achieves the highest mean accuracy and AUC within each cancer group.

Figures & tables

Figure 1: (a) Granularity mismatch between existing molecular encoders and sample-level proteomic states. (b) Cross-modal transfer is restricted to paired WSI–omics anchors. (c) Our framework leverages standalone molecular profiles through cross-modal knowledge distillation.

Figure 2: Overview of the three-stage framework. (a) PFE learns proteomic representations through virtual profile generation and expression-space multi-view pretraining. (b) ProSlide learns hierarchical WSI representations through protein-expression prediction. (c) Prot2Path distills proteomic relational knowledge into ProSlide through similarity distribution alignment over a shared sampled protein bank.

Model	Renal		Breast		Lung
	ACC	AUC	ACC	AUC	ACC	AUC
mean-pooling	57.82 $\pm$ 0.25	72.16 $\pm$ 0.80	67.74 $\pm$ 0.87	75.70 $\pm$ 1.01	59.61 $\pm$ 0.20	75.31 $\pm$ 0.20
FEATHER	54.95 $\pm$ 0.44	70.96 $\pm$ 1.01	66.79 $\pm$ 0.92	74.80 $\pm$ 0.82	60.22 $\pm$ 0.22	74.78 $\pm$ 0.26
GigaPath	56.04 $\pm$ 0.25	75.33 $\pm$ 0.72	67.19 $\pm$ 0.84	77.50 $\pm$ 1.08	58.93 $\pm$ 0.35	73.43 $\pm$ 0.27
PRISM	56.28 $\pm$ 0.36	72.72 $\pm$ 0.73	66.39 $\pm$ 0.88	74.88 $\pm$ 0.90	61.83 $\pm$ 0.16	74.86 $\pm$ 0.17
CARE	58.43 $\pm$ 0.31	75.24 $\pm$ 0.69	67.63 $\pm$ 0.67	75.07 $\pm$ 0.83	61.94 $\pm$ 0.20	76.99 $\pm$ 0.16

Table 1: Logistic-regression results averaged within the selected renal cell carcinoma (RCC), breast, and lung task groups. Each entry reports the mean of the four task-level means

\pm

the arithmetic mean of their reported standard deviations.

Model	Mask Results			External Mask Results			Distillation Results
Model	M1 R@1	M2 R@1	M3 R@1	M1 R@1	M2 R@1	M3 R@1	ACC	AUC
Protein-BERT	97.62	94.25	91.40	96.22	92.15	84.87	62.64	75.99
PFE & $B_{\mathrm{v}}=0$	97.42	93.89	89.90	95.35	91.71	83.35	63.21	76.94
PFE & $B_{\mathrm{v}}=32$	98.47	96.36	93.76	97.09	94.55	88.51	64.57	77.13
PFE & $B_{\mathrm{v}}=64$	98.92	96.92	95.17	96.58	95.05	88.44	64.68	77.70
PFE & $B_{\mathrm{v}}=128$	96.14	87.77	75.29	94.25	88.00	75.35	64.57	78.13

Table 2: Comparison of protein models on internal and external masked-protein retrieval tasks. The internal results are evaluated on 891 held-out TCGA samples, while the external results are evaluated using the combined 275-patient CPTAC gallery. All retrieval results are reported as the mean R@1 (%) over five runs.

S2-Sub	S2-Prot	S3-KD	Eff. rank	Renal		Breast		Lung
S2-Sub	S2-Prot	S3-KD	Eff. rank	ACC	AUC	ACC	AUC	ACC	AUC
✓	✗	✗	11.95	52.61	69.79	57.94	69.56	60.12	73.18
✗	✓	✗	232.92	58.07	75.86	69.22	76.69	61.34	75.39
✗	✗	✓	295.43	59.44	74.88	68.20	77.51	61.92	76.94
✓	✗	✓	12.12	52.32	70.26	58.34	72.48	61.11	74.41
✗	✓	✓	201.29	60.85	76.67	70.02	78.53	63.17	77.90

Table 3: Ablation study of supervision and knowledge distillation. S2-Sub and S2-Prot denote Stage II subtype supervision and protein-expression supervision, respectively. S3-KD denotes Stage III knowledge distillation. Eff. rank denotes the effective rank of representations on the CPTAC datasets.

Method	Target	Reference	Renal		Breast		Lung
Method	Target	Reference	ACC	AUC	ACC	AUC	ACC	AUC
Direct alignment	Cosine	Paired only	60.07	77.86	69.22	78.69	62.34	77.39
Hard matching	CE	All profiles	60.83	76.49	68.47	76.57	62.89	77.69
Paired-bank KD	Soft KL	Paired only	60.62	76.61	69.35	77.65	62.69	77.47
Full-bank KD	Soft KL	All profiles	60.33	76.90	70.14	78.69	62.60	77.77
Prot2Path	Soft KL	Sub-bank	60.85	76.67	70.02	78.53	63.17	77.90

Table 4: Ablation of Prot2Path components. We evaluate relational targets, standalone references, and sub-bank construction.

Figure 3: Visualization comparison of original WSIs and heatmaps. Heatmaps overlay patch aggregation weights for multiple-instance models or final-layer [CLS]-to-image-token attention weights for transformer-based models onto the corresponding WSIs.

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Resource	Records	Sample Domain	Protein Profile Type
TCGA	7,349	Patient tumors	Baseline tumor RPPA
TCPA	878	Cancer cell lines	Baseline cell-line RPPA
CCLE	899	Cancer cell lines	Baseline cell-line RPPA
CPPA	4,460	Perturbed cell lines	Treatment-induced $\Delta p$
Total	13,586	–	–

Appendix

Table S1: Composition of the proteomic cohort.

Panel Name	Standardized Target	Functional Annotation
ATM	ATM	DNA damage response
ATR	ATR	DNA damage response
BRCA2	BRCA2	Homologous recombination repair
P53	TP53	Tumor suppression / DNA damage response
RB	RB1	Cell-cycle regulation
BAP1C4	BAP1	Chromatin regulation / tumor suppression

Appendix

Table S2: Composition of the 38-protein panel. Total-protein and site-specific phosphorylation measurements are retained as distinct targets.

Component	Configuration	Description
Protein panel size	38	Shared protein vocabulary
Input sequence length	$39$	One [CLS] and 38 protein tokens
Value projection $g_{\mathrm{val}}$	$1\rightarrow 32\rightarrow 768$	Projects continuous abundance values
Missing-value embedding	$1\times 768$	Shared learnable value-mask embedding
Hidden dimension	768	Transformer token dimension
Transformer depth	7	Number of encoder blocks

Appendix

Table S3: Architecture configuration of the Proteomic Foundation Encoder (PFE). PFE represents each protein as the sum of a learnable identity embedding and a continuous abundance-value embedding, followed by Transformer-based contextual modeling.

Category	Hyperparameter	Value
Expression views	Profile-wide views	3
	Profile-wide view composition	1 clean + 2 posterized
	Subset-perturbed views	6
	Subset-perturbed Gaussian noise	$\sigma=0.05$
	Gaussian perturbation ratio	30% of observed proteins
Protein masking	Profile-wide student mask ratios	0.05 / 0.10 / 0.15

Appendix

Table S4: Self-supervised pretraining configuration of PFE. The model is trained using Distributed Data Parallel (DDP) on four NVIDIA RTX 4090 GPUs.

Model	Primary Input	Learning Objective	Representation / Output
ESM-2 [ Lin et al., 2022 ]	Amino-acid sequences	Masked amino-acid prediction	Residue and protein embeddings
AlphaFold 2 [ Jumper et al., 2021 ]	Sequences, alignments, and structural templates	Structure prediction	Three-dimensional protein structures
AlphaFold 3 [ Abramson et al., 2024 ]	Biomolecular sequences and ligand identities	Diffusion-based structure prediction	Three-dimensional biomolecular complexes
scPROTEIN [ Li et al., 2024 ]	Single-cell peptide and protein measurements	Graph contrastive learning and uncertainty estimation	Cell embeddings
scpFormer [ Zhou et al., 2026 ]	Single-cell protein abundance and sequence-derived embeddings	Multi-objective masked expression reconstruction	Cell and contextualized protein embeddings
VirTues [ Wenckstern et al., 2026 ]	Multiplex protein images and sequence-derived marker embeddings	Masked image reconstruction	Marker, cell, niche, and tissue representations

Appendix

Table S5: Comparison of representative protein modeling approaches. We summarize primary inputs, principal learning objectives, and representations or outputs.

Parameter	Configuration
Patch encoder	CONCH v1.5 (frozen)
Patch feature dimension	768
Region grid	$4\times 4$ patches
RSA depth	3
RSA attention heads	8
RSA MLP ratio	2

Appendix

Table S6: Architectural configuration of ProSlide.

Parameter	Configuration
Proteomic teacher	Frozen PFE
Teacher representation	Final [CLS] output after LayerNorm
Representation dimension	768
Candidate pool size	12,695 real profiles
Sub-bank size	$K_{\mathrm{sub}}=1{,}594$
Paired references	All 1,338 paired training profiles

Appendix

Table S7: Implementation configuration of Prot2Path.

Task	Cohort(s)	Patients ( $N_{p}$ )	WSIs ( $N_{s}$ )
Task 1	MUT-HETRCC	1292	1292
Task 2	MUT-HETRCC	1292	1292
Task 3	CPTAC-CCRCC	103	245
Task 4	MUT-HETRCC + CPTAC-CCRCC	1395	1537
Task 5	CPTAC-BRCA	97	106
Task 6	CPTAC-BRCA	82	89

Appendix

Table S8: Task-specific cohort sizes for downstream evaluation.

Model	Patch Encoder	Slide Aggregation	Pretraining Scale	Pretrain Sup.
CARE	CONCH v1.5	CARE	34,277 WSIs	iBoT & Clip
TITAN	CONCH v1.5	ViT	335,645 WSIs	iBoT & Coca
PRISM	Virchow	Perceiver	587,196 WSIs	Coca
GigaPath	GigaPath	LongNet	171,189 WSIs	MAE
FEATHER	CONCH v1.5	ABMIL	24,000 WSIs	Supervised classification
ProSlide	CONCH v1.5	Region ViT	12,695 protein profiles	Protein supervision &

Appendix

Table S9: Comparison of representative slide-level pathology foundation models. We compare the patch encoder, whole-slide aggregation strategy, pretraining scale, and supervision of the pathology foundation models considered in this work.

Task	Metric	Mean-pooling	FEATHER	GigaPath	PRISM	CARE	TITAN	ProSlide
Task 1	ACC	54.52 $\pm$ 0.04	54.88 $\pm$ 0.07	55.98 $\pm$ 0.04	55.93 $\pm$ 0.11	56.72 $\pm$ 0.02	56.29 $\pm$ 0.07	61.65 $\pm$ 0.20
Task 1	AUC	70.32 $\pm$ 0.13	73.41 $\pm$ 0.10	72.71 $\pm$ 0.05	72.29 $\pm$ 0.10	73.54 $\pm$ 0.08	72.56 $\pm$ 0.06	76.09 $\pm$ 0.07
Task 2	ACC	60.95 $\pm$ 0.04	54.74 $\pm$ 0.03	61.44 $\pm$ 0.13	57.55 $\pm$ 0.15	61.44 $\pm$ 0.07	59.83 $\pm$ 0.14	62.87 $\pm$ 0.16
Task 2	AUC	84.84 $\pm$ 0.06	84.28 $\pm$ 0.05	86.57 $\pm$ 0.01	86.49 $\pm$ 0.01	88.89 $\pm$ 0.02	85.98 $\pm$ 0.01	86.89 $\pm$ 0.03
Task 3	ACC	57.50 $\pm$ 0.83	58.14 $\pm$ 1.61	55.18 $\pm$ 0.65	53.98 $\pm$ 1.13	58.70 $\pm$ 1.03	61.20 $\pm$ 1.43	60.39 $\pm$ 1.14
Task 3	AUC	62.61 $\pm$ 2.92	58.23 $\pm$ 3.68	60.08 $\pm$ 2.79	59.52 $\pm$ 2.78	65.83 $\pm$ 2.61	73.16 $\pm$ 2.08	67.69 $\pm$ 2.31

Appendix

Table S10: Logistic-regression results on the selected 12-task set. Values are mean

\pm

standard deviation. The best and second-best means within each task-metric row are shown in bold and underlined, respectively.

Figure S1: Data and memory efficiency. ProSlide achieves strong performance with limited slide-level pretraining data (a) and low GPU memory usage during inference (b).

Figure S2: PFE protein identity embeddings. Pairwise cosine similarities among 38 learned protein identity embeddings. Both axes follow the same clustering order, with warmer colors indicating higher similarity.

Explore similar work

MOOZY: A Patient-First Foundation Model for Computational Pathology
Mar 27, 2026Yousef Kotp, Vincent Quoc-Huy Trinh, Christopher Pal +1Computational Pathology Self-Supervised Pre-Training
CSSE, Concordia University, Montréal, Canada · Mila – Québec AI Institute, Montréal, Canada · CHUM, Université de Montréal, Montréal, Canada +4
CARE: A Molecular-Guided Foundation Model with Adaptive Region Modeling for Whole Slide Image Analysis
Feb 25, 2026Di Zhang, Zhangpeng Gong, Xiaobo Pang +14Pathology Foundation Models Whole-Slide Image Analysis
Xi’an Jiaotong University · University of Cambridge · KingMed +2
From Patches to Patients: A study of the tile-to-slide performance transferability in Digital Pathology
Jun 9, 2026Sofiène Boutaj, Leo Fillioux, Maria Vakalopoulou +2Pathology Foundation Models Medical Image Benchmarks
Université Paris-Saclay, CentraleSupélec, Gustave Roussy, INSERM, IHU PRISM, Cancer Data Science Unit, France · Université Paris-Saclay, CentraleSupélec, MICS Laboratory, France

Mar 27, 2026J/K move · Enter open · S save

MOOZY: A Patient-First Foundation Model for Computational Pathology

Yousef Kotp, Vincent Quoc-Huy Trinh, Christopher Pal, Mahdi S. Hosseini

CSSE, Concordia University, Montréal, Canada · Mila – Québec AI Institute, Montréal, Canada · CHUM, Université de Montréal, Montréal, Canada+4

Computational Pathology Self-Supervised Pre-Training Pathology Foundation Models Clinical Prediction Whole-Slide Image Analysis

Open paper AskPDF arXiv

Computational pathology needs whole-slide image (WSI) foundation models that transfer across diverse clinical tasks, yet current approaches remain largely slide-centric, often depend on private data and expensive paired-report supervision, and do not explicitly model relationships among multiple slides from the same patient. We present MOOZY, a patient-first pathology foundation model in which the patient case, not the individual slide, is the core unit of representation. MOOZY explicitly models dependencies across all slides from the same patient via a case transformer during pretraining, combining multi-stage self-supervision with scaled low-cost task supervision. In Stage 1, we pretrain a vision-only slide encoder on 77,134 public slide feature grids using masked self-distillation. In Stage 2, we align these representations with clinical semantics using a case transformer and multi-task supervision over 333 tasks from 56 public datasets, including 205 classification and 128 survival tasks across four endpoints. Across sixteen held-out tasks, MOOZY improves macro weighted F1, balanced accuracy, and macro weighted ROC-AUC relative to PRISM by +4.19%, +7.93%, and +6.95%, respectively. MOOZY is also parameter efficient with 85.77M parameters, 14

\times

smaller than GigaPath. These results suggest that patient-level pretraining yields transferable embeddings, providing a path toward scalable patient-first histopathology foundation models.

Abstract

Explore similar work

MOOZY: A Patient-First Foundation Model for Computational Pathology

CARE: A Molecular-Guided Foundation Model with Adaptive Region Modeling for Whole Slide Image Analysis

From Patches to Patients: A study of the tile-to-slide performance transferability in Digital Pathology