Generalization of Self-Supervised Vision Transformers for Protein Localization Across Microscopy Domains

TL;DR

This study demonstrates that self-supervised pretrained Vision Transformers can effectively transfer across different microscopy domains for protein localization, achieving high accuracy even with limited labeled data.

Contribution

It shows that domain-specific self-supervised pretraining enhances transferability of Vision Transformers across microscopy datasets, outperforming models trained directly on target data.

Findings

HPA-pretrained model achieved the highest macro F1-score of 0.8221.

All pretrained models transferred well across domains.

Domain-relevant SSL representations generalize effectively to related microscopy datasets.

Abstract

Task-specific microscopy datasets are often too small to train deep learning models that learn robust feature representations. Self-supervised learning (SSL) can mitigate this by pretraining on large unlabeled datasets, but it remains unclear how well such representations transfer across microscopy domains with different staining protocols and channel configurations. We investigate the cross-domain transferability of DINO-pretrained Vision Transformers for protein localization on the OpenCell dataset. We generate image embeddings using three DINO backbones pretrained on ImageNet-1k, the Human Protein Atlas (HPA), and OpenCell, and evaluate them by training a supervised classification head on OpenCell labels. All pretrained models transfer well, with the microscopy-specific HPA-pretrained model achieving the best performance (mean macro $F_1$-score = 0.8221 $\pm$ 0.0062), slightly outperforming a DINO model trained directly on OpenCell (0.8057 $\pm$ 0.0090). These results highlight the value of large-scale pretraining and indicate that domain-relevant SSL representations can generalize effectively to related but distinct microscopy datasets, enabling strong downstream performance even when task-specific labeled data are limited.