C3R: Channel Conditioned Cell Representations for unified evaluation in microscopy imaging

Muhammad Awais; Sara Atito; Syed Sameed Husain; Umar Marikkar

REVIEW 3 major objections 5 minor 51 references

C3R: Channel Conditioned Cell Representations for unified evaluation in microscopy imaging

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims C3R, a context-concept channel split with a branched encoder and masked context distillation, enables training-free zero-shot evaluation across immunofluorescence datasets with different channel configurations, matching…

desk verdict The context-concept idea is a real contribution, but the paper's zero-shot OOD claims rest on test-set selection and noise-level margins; the ID results are solid and it deserves revision, not rejection. read the letter →

arxiv 2505.18745 v1 pith:6LLKJA4Z submitted 2025-05-24 cs.CV cs.LGq-bio.QM

Umar Marikkar , Syed Sameed Husain , Muhammad Awais , Sara Atito This is my paper · ORCID

classification cs.CVcs.LGq-bio.QM

keywords immunohistochemistrychannel-adaptiveencoderself-supervisedlearningzero-shotout-of-distributionevaluationcontext-conceptchannelsplitcellrepresentationvisiontransformermicroscopyimaging

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Immunofluorescent cell images come with different numbers of channels depending on the staining protocol, so most models must be re-trained for each dataset. This paper claims that the channels themselves fall into two roles: context channels (nucleus, endoplasmic reticulum, microtubules) that stay visually consistent across cells and datasets, and concept channels that carry the experiment-specific signal such as which protein is stained or how a drug perturbs the cell. On that split the paper builds C3R, a branched encoder with a masked distillation training objective, and claims that one encoder pre-trained once on the four-channel source data can be evaluated directly on datasets with different channel configurations, matching or beating baselines that were re-trained for each target. A stripped-down grouped-stem version of the idea also scores highest among stem networks on a multi-dataset microscopy benchmark. If the claim transfers, cross-dataset evaluation in immunofluorescence no longer requires dataset-specific adaptation.

What carries the argument

The load-bearing object is the context-concept channel split instantiated as a branched Vision Transformer: context channels (nucleus, endoplasmic reticulum, microtubules) go through their own convolutional stem (initial per-channel feature extractor) and branched layers, concept channels through a second stem and branch, and the two group-wise outputs are concatenated and passed to a shared encoder. The same split drives Masked Context Distillation, a self-supervised objective in which the student sees a randomly reduced subset of context channels and must produce the same representation as a teacher that always sees the full context; the KL divergence between their projection heads is the loss. The architecture carries the transfer claim because the context branch must encode stable structure while the concept branch encodes variable signal, so on a new dataset the same context channels can anchor evaluation without retraining.

What would settle it

Run zero-shot retrieval with the C3R encoder on a new immunofluorescence dataset whose presumed context channel (for example, the endoplasmic-reticulum stain) has cluster parity and entropy close to the concept channels' values; if accuracy drops to roughly the level of a randomly initialized encoder or far below a per-dataset re-trained baseline, the context-concept assumption is not transferable and the unified-evaluation claim fails. The paper itself records one warning of this kind (the ER channel's parity 0.489 and entropy 1.496 on the target dataset), so the decisive test is a dataset where that pattern is the rule rather than an exception.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that treating context channels as a stable reference and concept channels as the variable of interest, instead of treating every channel symmetrically, is what makes cell-image representations transfer across staining configurations. The Context-Concept Encoder passes context and concept channels through separate stems and separate branched encoder layers, normalizes each channel, merges the two group representations, and then runs a shared encoder; Masked Context Distillation trains the student to reproduce a full-context teacher's representation while the student randomly loses one or more context channels. The paper reports that this outperforms the in-distribution protein-localization benchmarks and matches or exceeds the out-of-distribution drug-perturbation retrieval baselines without any adaptation, that the out-of-distribution gain comes largely from the architecture rather than from the distillation loss, and that even the naive grouped-stem version scores highest (0.626) among stem networks on the multi-dataset microscopy benchmark.

Load-bearing premise

The whole transfer story rests on the assumption that a target dataset's context channels can be identified by hand and that those channels really are visually consistent across datasets; if a presumed context channel turns out to be highly variable, the training-free evaluation loses its anchor and the method falls back to the baselines it claims to beat.

Editorial extensions

If this is right

A single pre-trained checkpoint can be evaluated on unseen datasets with different channel counts and configurations, removing the per-dataset retraining cost that current channel-adaptive models incur.
The grouped-stem version of the core idea beats more elaborate channel-adaptive stems on a multi-dataset benchmark, which suggests the context-concept grouping itself, not the conditioning machinery, is the main driver.
Masked context distillation improves in-distribution protein-localization accuracy by roughly two percent and makes the representation robust to dropping a context channel at inference time.
The out-of-distribution gains come from the branched architecture and instance normalization, while the distillation loss mainly helps in-distribution performance.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A natural extension the paper leaves implicit is automatic discovery of the context-concept split: the parity and entropy statistics used to validate the assumption could be turned into a data-driven rule for assigning channels on an unseen dataset, removing the need for hand-labeling.
If the split holds for other multiplexed assays, such as spatial transcriptomics or multiplexed histopathology, the same two-stem strategy could supply training-free evaluation across platforms with different marker panels, not just fluorescence microscopy.
Because the paper finds the distillation loss benefits mostly in-distribution, a reasonable next experiment is to condition only the concept branch on context rather than distilling the whole global representation; that would reveal whether the transfer mechanism is the architectural grouping or the distillation objective.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The context–concept split of IHC channels is a genuinely useful organizing principle, and the ID results are the strongest part of this paper. The CCE architecture is a modest extension, but the combination with masked context distillation is new, and the HPA gains are consistent and come with error bars (e.g., ViT-B 31-loc 0.548 vs 0.515 baseline). The CHAMMI result with a naive grouped stem is also nice: they used the official benchmark code and beat four established stem networks, even if the CPS margin over the best baseline is only 0.003. The paper is commendably honest in its limitations section, explicitly saying MCD's gains don't transfer to OOD and acknowledging the context-concept assumption may not hold for other datasets.

The soft spots are in the OOD evaluation, and they are load-bearing for the headline claim. Table 1 shows ViT-S C3R mAP 0.354 vs baseline 0.355—essentially a tie, actually slightly below. The JUMP-CP protocol (Section B.2) selects the best post-processing configuration (PCA/ZCA, MAD/z-score/None) on the test set, for every method. That is per-method test-set selection, making each reported OOD number an optimistic upper bound, and the comparison can flip when a single configuration is fixed in advance. The OOD columns have no error bars or significance tests. Table 3 shows MCD actually lowers OOD metrics (CCE-only 0.444 vs C3R 0.436 for ViT-S), which the paper admits. So the zero-shot OOD advantage is not established by the reported evidence. Also, the ER channel in JUMP-CP doesn't behave like a context channel in their own parity/entropy analysis (Table 5), weakening the transfer assumption for that dataset. These are proportional concerns: the idea might still work, but the evidence is mixed.

Who this is for: researchers in multi-channel microscopy representation learning and anyone thinking about cross-dataset evaluation in bioimaging. I'd send it to review because the context-concept principle is new enough and the ID results are solid, but the authors need to fix the OOD protocol. Concretely: rerun the OOD evaluation with a held-out post-processing configuration, report variances and significance, and soften or remove the claim about OpenCell and Bridge2AI until those datasets are actually evaluated. With that revision, this would be a useful contribution.

Referee Report

3 major / 5 minor

Summary. The paper proposes C3R, a framework for immunohistochemistry (IHC) cell-image representation learning built on a context/concept dichotomy of imaging channels. Context channels (e.g., Nucleus, ER, Microtubules) are treated as structural references, while concept channels carry experiment-specific signal. The method has two components: a Context-Concept Encoder (CCE) that processes context and concept channel groups separately before merging, and a Masked Context Distillation (MCD) training strategy that drops context channels in the student while the teacher sees the full context. The authors pre-train ViT-S and ViT-B models on HPA with iBOT plus a SubCell-style antibody loss, evaluate ID performance on HPA (19- and 31-location mAP), evaluate zero-shot OOD performance on JUMP-CP (mAP and kNN retrieval), and also insert a grouped stem into the CHAMMI benchmark. The central claims are that C3R outperforms existing benchmarks on both ID and OOD tasks and enables training-free cross-dataset evaluation.

Significance. The context/concept perspective is a useful organizing principle for multi-channel microscopy, and the ID results are consistent and carry error bars: for example, ViT-B C3R reaches 0.548 HPA 31-loc mAP versus 0.515 for the re-trained baseline (Table 1). The grouped stem also achieves the highest CHAMMI Performance Score (0.626) among the compared stem networks, albeit by a small margin. The paper includes a useful ablation chain (Table 3), a DINOv2/UMAP-based validation of channel grouping (Appendix A.1), and a link to an anonymous code repository. If the zero-shot OOD claim were fully established, this would be a meaningful step toward cross-dataset generalization in IHC imaging. However, as reported, the zero-shot OOD evidence is not yet strong enough to support the abstract's claim that C3R outperforms existing benchmarks on both ID and OOD tasks, because the JUMP-CP evaluation protocol selects post-processing on the test set, the OOD margins are at noise level, and one of the table rows (ViT-S C3R mAP) is below the re-trained baseline.

major comments (3)

[B.2, Table 1] The zero-shot JUMP-CP results are obtained by selecting the best of six post-processing configurations {PCA, ZCA} x {MAD, z-score, None} on the test set itself, as stated in B.2: the authors 'report results using the configuration that yields the highest performance.' This is test-set selection, and because the selection is performed independently per method, the Table 1 OOD comparisons are optimistic upper bounds rather than fair evaluations. The problem is compounded by the absence of error bars or significance tests on the OOD columns: the ViT-S C3R mAP is 0.354 versus 0.355 for the re-trained baseline, and the ViT-B C3R mAP gain is only 0.008 over baseline. To support the central 'matches or exceeds' OOD claim, the post-processing configuration should be fixed in advance, selected on a held-out validation split (e.g., on held-out plates or a subset of perturbations), or reported with standard deviations over replicates; otherwise the apparent OOD superiority may be an artifact of the selection protocol.
[4.2, Table 3] The component analysis shows that the full C3R method (CCE + MCD) does not consistently improve OOD performance over CCE alone. For ViT-S, CCE has JUMP-CP OOD averages of mAP 0.358 and kNN 0.530, while C3R has mAP 0.354 and kNN 0.518, i.e., MCD reduces both OOD metrics and also pushes mAP below the baseline's 0.355. For ViT-B, C3R OOD mAP is 0.363 versus 0.358 for CCE, but the kNN values are 0.530 versus 0.532, again a mixed outcome. The paper's own limitations section acknowledges that MCD gains do not translate to OOD. Therefore the abstract's claim that 'C3R outperforms existing benchmarks on both ID and OOD tasks' is too strong as written: the OOD advantage, where present, is attributable to the CCE architecture rather than to the full C3R framework, and even that advantage is not uniform across metrics or backbones. The claims and the abstract should be revised to distinguish ID gains from the more tentative OOD results.
[A.1, Table 5] The zero-shot deployment of C3R requires that a target dataset's context channels be identifiable, and the authors assign these groups by hand (Section 3.2, Fig. 1). The quantitative validation in Appendix A.1 finds that the ER channel on JUMP-CP, which is treated as a context channel, has parity 0.489 and entropy 1.496, values inconsistent with the expected context behavior seen for the other context channels (e.g., Nucleus parity 0.943, entropy 0.372). If the context/concept assignment is wrong for a target dataset, the claimed training-free OOD evaluation loses its foundation. Because the paper's contribution is precisely a unified evaluation method for datasets with different channel configurations, the manual assignment of groups for JUMP-CP is load-bearing rather than a minor implementation detail. The authors should either provide an automatic, validated procedure for identifying context versus concept channels on a new dataset, or explicitly restrict the OOD claims to datasets whose group assignment is known in advance and demonstrate robustness by testing alternative assignments for JUMP-CP.

minor comments (5)

[Eq. (8)] The condition on the number of dropped channels is written as '1 <= c < C1' but then restated as 'where 0 <= c < C1'; the second inequality contradicts the first and should be corrected.
[B.2] The text states that JUMP-CP images are resized to '112 x 122' pixels; this is likely a typo for '112 x 112' and should be corrected.
[3.1] The sentence 'Based on the structural coherence of context channels, Using MCD, we encourage...' is grammatically incomplete and should be rewritten.
[4.2, Fig. 3] The group-switching experiment in Fig. 3 reports single values without error bars or multiple seeds; adding variance estimates would make the claimed drop in performance more convincing.
[A.1] The DINOv2/UMAP validation is performed on N=1000 instances without reporting the number of seeds or the variance of the parity/entropy estimates; a small bootstrap or repeated-clustering analysis would strengthen the claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the C3R OOD evaluation is genuine zero-shot transfer from an HPA-pretrained encoder, and the context-concept prior is independently validated rather than fitted to the target dataset.

full rationale

The central derivation chain is self-contained. The context-concept grouping is introduced as a biological prior (Sec. 3.1, Eq. 1, Fig. 1), grounded in external references [13,14] and independently probed in Appendix A.1 using DINOv2 features and K-Means parity/entropy metrics, not fitted to C3R's JUMP-CP performance. CCE and MCD are trained exclusively on HPA (Sec. 3.2, B.2), and the JUMP-CP evaluation uses the resulting frozen encoder with no retraining or adaptation; Table 1 labels baselines that were retrained for JUMP-CP, whereas C3R is not. No equation defines the target metric in terms of the model's own fitted parameters, and no load-bearing result is imported from the authors' prior work. The authors explicitly acknowledge the main limitation: OOD transfer depends on the context-concept assumption holding for the target dataset, and MCD's ID gains do not translate to OOD (Sec. 5). The post-processing configuration selection in B.2 is a test-set selection concern about evaluation validity, not a circularity reduction: it does not make the reported OOD metric equal to the input by construction. Accordingly, the paper is assigned a circularity score of 0.

Assumptions & free parameters 5 free parameters · 3 assumptions · 0 invented entities

The 'derivation' here is a network design plus a training objective; the load-bearing hand-chosen ingredients are the context/concept grouping, branch and aggregation configuration, MCD sampling, and test-time post-processing selection. No new physical entities or fitted constants in the physics sense are introduced.

free parameters (5)

Context/concept group assignments = HPA: context={Microtubules, Nucleus, ER}, concept={Protein}; JUMP-CP: context={Nucleus, ER}, concept={Mitochondria…
Hand-chosen per dataset (Fig 1, Sec 3.2). Load-bearing: the encoder and MCD are defined relative to these groups, and flipping them degrades performance (Fig 3).
Branch encoder depth (fc1, fc2) = 2 layers per branch; 10-11 shared layers (parameter-normalized)
Selected by validation sweep in Appendix A.2 (Fig 5); affects representational granularity and merger timing.
Aggregation variant = pre-aggregation without MCD; post-aggregation with MCD
Chosen empirically (Table 6); changes whether branches encode group-level or channel-level features.
MCD context channel sampling distribution = c uniformly sampled from {0,1,2}
Optimal sampling found by ablation in Appendix A.4 (Table 7).
JUMP-CP post-processing configuration = best of {PCA,ZCA} x {MAD-normalization, z-score, None}
Selected on test data ('We report results using the configuration that yields the highest performance'), introducing selection bias; Appendix B.

assumptions (3)

domain assumption IHC channels can be partitioned into context and concept groups, where context channels are structurally consistent across instances and datasets.
Core premise of the paper, introduced in Sec 1 and Fig 1 and used in CCE (Sec 3.1) and MCD. Partially validated in Appendix A.1, but ER on JUMP-CP violates the expected pattern (Table 5).
standard math iBOT self-supervised objective with the SubCell antibody-supervised contrastive loss is a suitable base for learning cell representations.
Taken from prior work [21,5] and used as the training backbone without modification; not derived in this paper.
domain assumption Mean pooling across channels (pre or post aggregation) preserves enough information for the downstream tasks.
Design choice in Eqs 3-5; not theoretically justified, only empirically validated.

how reviews work

0 comments

Cite this review

Pith. "Pith review of C3R: Channel Conditioned Cell Representations for unified evaluation in microscopy imaging." pith.science (2026). https://pith.science/paper/6LLKJA4Z

@misc{pith2026250518745,
  author       = {Pith},
  title        = {Pith review of: C3R: Channel Conditioned Cell Representations for unified evaluation in microscopy imaging},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6LLKJA4Z}},
  note         = {Machine review of arXiv:2505.18745}
}

read the original abstract

Immunohistochemical (IHC) images reveal detailed information about structures and functions at the subcellular level. However, unlike natural images, IHC datasets pose challenges for deep learning models due to their inconsistencies in channel count and configuration, stemming from varying staining protocols across laboratories and studies. Existing approaches build channel-adaptive models, which unfortunately fail to support out-of-distribution (OOD) evaluation across IHC datasets and cannot be applied in a true zero-shot setting with mismatched channel counts. To address this, we introduce a structured view of cellular image channels by grouping them into either context or concept, where we treat the context channels as a reference to the concept channels in the image. We leverage this context-concept principle to develop Channel Conditioned Cell Representations (C3R), a framework designed for unified evaluation on in-distribution (ID) and OOD datasets. C3R is a two-fold framework comprising a channel-adaptive encoder architecture and a masked knowledge distillation training strategy, both built around the context-concept principle. We find that C3R outperforms existing benchmarks on both ID and OOD tasks, while a trivial implementation of our core idea also outperforms the channel-adaptive methods reported on the CHAMMI benchmark. Our method opens a new pathway for cross-dataset generalization between IHC datasets, without requiring dataset-specific adaptation or retraining.

Figures

Figures reproduced from arXiv: 2505.18745 by the authors.

**Figure 2.** Overview of C3R. (a) Context-Concept Encoder: The input channels are separated into [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Effects of group switching assignments. The experiments were carried out for JUMP-CP using ViT-S without MCD and d layers per branch [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Visualization of channel-wise features in 2D space using UMAP on (a) HPA and (b) [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]

**Figure 5.** Figure 5: Comparison of performance impact from (a) pre-merging and (b) post-merging depths on [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]

**Figure 6.** Figure 6: (a) Pre-aggregation. The channels are passed separately through the convolutional stem, [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]

**Figure 7.** Figure 7: (a) Cosine similarity between full and sparse intermediate representations [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 31 canonical work pages

[1]

Cell painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes.Nature protocols, 11(9):1757–1774, 2016

Mark-Anthony Bray, Shantanu Singh, Han Han, Chadwick T Davis, Blake Borgeson, Cathy Hartland, Maria Kost-Alimova, Sigrun M Gustafsdottir, Christopher C Gibson, and Anne E Carpenter. Cell painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes.Nature protocols, 11(9):1757–1774, 2016

work page 2016
[2]

Cell painting predicts impact of lung cancer variants.Molecular biology of the cell, 33, 2022

Juan C Caicedo, John Arevalo, Federica Piccioni, Mark-Anthony Bray, Cathy L Hartland, Xiaoyun Wu, Angela N Brooks, Alice H Berger, Jesse S Boehm, Anne E Carpenter, et al. Cell painting predicts impact of lung cancer variants.Molecular biology of the cell, 33, 2022

work page 2022
[3]

Single-cell subcellular protein localisation using novel ensembles of diverse deep architectures

Syed Sameed Husain, Eng-Jon Ong, Dmitry Minskiy, Mikel Bober-Irizar, Amaia Irizar, and Miroslaw Bober. Single-cell subcellular protein localisation using novel ensembles of diverse deep architectures. Communications biology, 6(1):489, 2023

work page 2023
[4]

Unbiased single-cell morphology with self-supervised vision transformers.bioRxiv, 2023

Michael Doron, Théo Moutakanni, Zitong S Chen, Nikita Moshkov, Mathilde Caron, Hugo Touvron, Piotr Bojanowski, Wolfgang M Pernice, and Juan C Caicedo. Unbiased single-cell morphology with self-supervised vision transformers.bioRxiv, 2023

work page 2023
[5]

Subcell: Vision foundation models for microscopy capture single-cell biology.bioRxiv, pages 2024–12, 2024

Ankit Gupta, Zoe Wefers, Konstantin Kahnert, Jan N Hansen, Will Leineweber, Anthony Cesnik, Dan Lu, Ulrika Axelsson, Frederic Ballllosera Navarro, Theofanis Karaletsos, et al. Subcell: Vision foundation models for microscopy capture single-cell biology.bioRxiv, pages 2024–12, 2024

work page 2024
[6]

Self-supervision advances morphological profiling by unlocking powerful image representations.Scientific Reports, 15(1):4876, 2025

Vladislav Kim, Nikolaos Adaloglou, Marc Osterland, Flavio M Morelli, Marah Halawa, Tim König, David Gnutt, and Paula A Marin Zapata. Self-supervision advances morphological profiling by unlocking powerful image representations.Scientific Reports, 15(1):4876, 2025

work page 2025
[7]

Berg, and Li Fei-Fei

Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg, and Li Fei-Fei. ImageNet Large Scale Visual Recognition Challenge.International Journal of Computer Vision (IJCV), 115(3):211–252,
[8]

A subcellular map of the human proteome

Peter J Thul, Lovisa Åkesson, Mikaela Wiking, Diana Mahdessian, Aikaterini Geladaki, Hammou Ait Blal, Tove Alm, Anna Asplund, Lars Björk, Lisa M Breckels, et al. A subcellular map of the human proteome. Science, 356(6340):eaal3321, 2017

work page 2017

Show all 51 references

[9]

Jump cell painting dataset: morphological impact of 136,000 chemical and genetic perturbations.BioRxiv, pages 2023–03, 2023

Srinivas Niranj Chandrasekaran, Jeanelle Ackerman, Eric Alix, D Michael Ando, John Arevalo, Melissa Bennion, Nicolas Boisseau, Adriana Borowa, Justin D Boyd, Laurent Brino, et al. Jump cell painting dataset: morphological impact of 136,000 chemical and genetic perturbations.Bi...

2023
[10]

Chammi: A benchmark for channel-adaptive models in microscopy imaging.Advances in Neural Information Processing Systems, 36:19700–19713, 2023

Zitong Sam Chen, Chau Pham, Siqi Wang, Michael Doron, Nikita Moshkov, Bryan Plummer, and Juan C Caicedo. Chammi: A benchmark for channel-adaptive models in microscopy imaging.Advances in Neural Information Processing Systems, 36:19700–19713, 2023

2023
[11]

Channel vision transformers: An image is worth 1 x 16 x 16 words

Yujia Bao, Srinivasan Sivanandan, and Theofanis Karaletsos. Channel vision transformers: An image is worth 1 x 16 x 16 words. InThe Twelfth International Conference on Learning Representations, 2024

2024
[12]

Enhancing feature diversity boosts channel-adaptive vision transformers

Chau Pham and Bryan Plummer. Enhancing feature diversity boosts channel-adaptive vision transformers. Advances in Neural Information Processing Systems, 37:89782–89805, 2024

2024
[13]

Dira: Discriminative, restorative, and adversarial learning for self-supervised medical image analysis

Fatemeh Haghighi, Mohammad Reza Hosseinzadeh Taher, Michael B Gotway, and Jianming Liang. Dira: Discriminative, restorative, and adversarial learning for self-supervised medical image analysis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognitio...

2022
[14]

Contextual vision transformers for robust representation learning

Yujia Bao and Theofanis Karaletsos. Contextual vision transformers for robust representation learning. arXiv preprint arXiv:2305.19402, 2023

2023 arXiv
[15]

Vitally consistent: Scaling biological representation learning for cell microscopy.arXiv preprint arXiv:2411.02572, 2024

Kian Kenyon-Dean, Zitong Jerry Wang, John Urbanik, Konstantin Donhauser, Jason Hartford, Saber Saberian, Nil Sahin, Ihab Bendidi, Safiye Celik, Marta Fay, et al. Vitally consistent: Scaling biological representation learning for cell microscopy.arXiv preprint arXiv:2411.02572, 2024

2024 arXiv
[16]

De Lorenci, Seung Eun Yi, Théo Moutakanni, Piotr Bojanowski, camille couprie, Juan C Caicedo, and Wolfgang Maximilian Anton Pernice

Alice V . De Lorenci, Seung Eun Yi, Théo Moutakanni, Piotr Bojanowski, camille couprie, Juan C Caicedo, and Wolfgang Maximilian Anton Pernice. Scaling channel-invariant self-supervised learning, 2025

2025
[17]

Masked autoencoders for microscopy are scalable learners of cellular biology

Oren Kraus, Kian Kenyon-Dean, Saber Saberian, Maryam Fallah, Peter McLean, Jess Leung, Vasudev Sharma, Ayla Khan, Jia Balakrishnan, Safiye Celik, et al. Masked autoencoders for microscopy are scalable learners of cellular biology. InProceedings of the IEEE/CVF Conference on Co...

2024
[18]

A simple framework for contrastive learning of visual representations

Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. A simple framework for contrastive learning of visual representations. InInternational conference on machine learning, pages 1597–1607. PmLR, 2020

2020
[19]

Emerging properties in self-supervised vision transformers

Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerging properties in self-supervised vision transformers. InProceedings of the IEEE/CVF international conference on computer vision, pages 9650–9660, 2021

2021
[20]

Masked autoencoders are scalable vision learners

Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, and Ross Girshick. Masked autoencoders are scalable vision learners. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 16000–16009, 2022

2022
[21]

ibot: Image bert pre-training with online tokenizer.arXiv preprint arXiv:2111.07832, 2021

Jinghao Zhou, Chen Wei, Huiyu Wang, Wei Shen, Cihang Xie, Alan Yuille, and Tao Kong. ibot: Image bert pre-training with online tokenizer.arXiv preprint arXiv:2111.07832, 2021

2021 arXiv
[22]

Microsnoop: A generalist tool for microscopy image representation.The Innovation, 5(1):100541, 2024

Dejin Xun, Rui Wang, Xingcai Zhang, and Yi Wang. Microsnoop: A generalist tool for microscopy image representation.The Innovation, 5(1):100541, 2024. ISSN 2666-6758. doi: https://doi.org/10.1016/j.xinn. 2023.100541

2024
[23]

Generalizable morphological profiling of cells by interpretable unsupervised learning

Rashmi Sreeramachandra Murthy, Shobana V Stassen, Dickson MD Siu, Michelle CK Lo, Gwinky GK Yip, and Kevin K Tsia. Generalizable morphological profiling of cells by interpretable unsupervised learning. bioRxiv, pages 2024–09, 2024

2024
[24]

Self-supervised deep learning encodes high-resolution features of protein subcellular localization.Nature methods, 19(8): 995–1003, 2022

Hirofumi Kobayashi, Keith C Cheveralls, Manuel D Leonetti, and Loic A Royer. Self-supervised deep learning encodes high-resolution features of protein subcellular localization.Nature methods, 19(8): 995–1003, 2022

2022
[25]

Weakly supervised set-consistency learning improves morphological profiling of single-cell images

Heming Yao, Phil Hanslovsky, Jan-Christian Huetter, Burkhard Hoeckendorf, and David Richmond. Weakly supervised set-consistency learning improves morphological profiling of single-cell images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition...

2024
[26]

Bidcell: Biologically-informed self-supervised learning for segmentation of subcellular spatial transcriptomics data.Nature Communications, 15(1):509, 2024

Xiaohang Fu, Yingxin Lin, David M Lin, Daniel Mechtersheimer, Chuhan Wang, Farhan Ameen, Shila Ghazanfar, Ellis Patrick, Jinman Kim, and Jean YH Yang. Bidcell: Biologically-informed self-supervised learning for segmentation of subcellular spatial transcriptomics data.Nature Co...

2024
[27]

Pedro Herruzo, Aleksandra Gruca, Llorenç Lliso, Xavier Calbet, Pilar Rípodas, Sepp Hochreiter, Michael Kopp, and David P. Kreil. High-resolution multi-channel weather forecasting – first insights on transfer learning from the weather4cast competitions 2021. In2021 IEEE Interna...

2021
[28]

Climax: A foundation model for weather and climate.arXiv preprint arXiv:2301.10343, 2023

Tung Nguyen, Johannes Brandstetter, Ashish Kapoor, Jayesh K Gupta, and Aditya Grover. Climax: A foundation model for weather and climate.arXiv preprint arXiv:2301.10343, 2023

2023 arXiv
[29]

Chada-vit: Channel adaptive attention for joint representation learning of heterogeneous microscopy images

Nicolas Bourriez, Ihab Bendidi, Ethan Cohen, Gabriel Watkinson, Maxime Sanchez, Guillaume Bollot, and Auguste Genovesio. Chada-vit: Channel adaptive attention for joint representation learning of heterogeneous microscopy images. InProceedings of the IEEE/CVF Conference on Comp...

2024
[30]

Isolated channel vision transformers: From single-channel pretraining to multi-channel finetuning.arXiv preprint arXiv:2503.09826, 2025

Wenyi Lian, Joakim Lindblad, Patrick Micke, and Nataša Sladoje. Isolated channel vision transformers: From single-channel pretraining to multi-channel finetuning.arXiv preprint arXiv:2503.09826, 2025

2025
[31]

An image is worth 16x16 words: Transformers for image recognition at scale

Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at...

2021
[32]

Human protein atlas - single cell classification

Casper Winsnes, Emma Lundberg, Maggie, Phil Culliton, Trang Le, UAxelsson, and Wei Ouyang. Human protein atlas - single cell classification. https://kaggle.com/competitions/ hpa-single-cell-image-classification, 2021. Kaggle

2021
[33]

Three million images and morphological profiles of cells treated with matched chemical and genetic perturbations.Nature Methods, 21(6):1114–1121, 2024

Srinivas Niranj Chandrasekaran, Beth A Cimini, Amy Goodale, Lisa Miller, Maria Kost-Alimova, Nasim Jamali, John G Doench, Briana Fritchman, Adam Skepner, Michelle Melanson, et al. Three million images and morphological profiles of cells treated with matched chemical and geneti...

2024
[34]

Integrated intracellular organization and its variations in human ips cells.Nature, 613(7943):345–354, 2023

Matheus P Viana, Jianxu Chen, Theo A Knijnenburg, Ritvik Vasan, Calysta Yan, Joy E Arakaki, Matte Bailey, Ben Berry, Antoine Borensztejn, Eva M Brown, et al. Integrated intracellular organization and its variations in human ips cells.Nature, 613(7943):345–354, 2023

2023
[35]

A convnet for the 2020s

Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, and Saining Xie. A convnet for the 2020s. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11976–11986, 2022

2022
[36]

Learning implicitly recurrent cnns through parameter sharing.arXiv preprint arXiv:1902.09701, 2019

Pedro Savarese and Michael Maire. Learning implicitly recurrent cnns through parameter sharing.arXiv preprint arXiv:1902.09701, 2019

1902 arXiv
[37]

Dai, and Quoc V

David Ha, Andrew M. Dai, and Quoc V . Le. Hypernetworks. InInternational Conference on Learning Representations, 2017

2017
[38]

Sdae: Self-distillated masked autoencoder

Yabo Chen, Yuchen Liu, Dongsheng Jiang, Xiaopeng Zhang, Wenrui Dai, Hongkai Xiong, and Qi Tian. Sdae: Self-distillated masked autoencoder. InEuropean conference on computer vision, pages 108–124. Springer, 2022

2022
[39]

Opencell: Endogenous tagging for the cartography of human cellular organization.Science, 375(6585):eabi6983, 2022

Nathan H Cho, Keith C Cheveralls, Andreas-David Brunner, Kibeom Kim, André C Michaelis, Preethi Raghavan, Hirofumi Kobayashi, Laura Savy, Jason Y Li, Hera Canaj, et al. Opencell: Endogenous tagging for the cartography of human cellular organization.Science, 375(6585):eabi6983, 2022

2022
[40]

Cell maps for artificial intelligence: Ai-ready maps of human cell architecture from disease-relevant cell lines.bioRxiv, 2024

Timothy Clark, Jillian Mohan, Leah Schaffer, Kirsten Obernier, Sadnan Al Manir, Christopher P Churas, Amir Dailamy, Yesh Doctor, Antoine Forget, Jan Niklas Hansen, et al. Cell maps for artificial intelligence: Ai-ready maps of human cell architecture from disease-relevant cell...

2024
[41]

Multiscale vision transformers

Haoqi Fan, Bo Xiong, Karttikeya Mangalam, Yanghao Li, Zhicheng Yan, Jitendra Malik, and Christoph Feichtenhofer. Multiscale vision transformers. InProceedings of the IEEE/CVF international conference on computer vision, pages 6824–6835, 2021

2021
[42]

Hiera: A hierarchical vision transformer without the bells-and-whistles

Chaitanya Ryali, Yuan-Ting Hu, Daniel Bolya, Chen Wei, Haoqi Fan, Po-Yao Huang, Vaibhav Aggarwal, Arkabandhu Chowdhury, Omid Poursaeed, Judy Hoffman, et al. Hiera: A hierarchical vision transformer without the bells-and-whistles. InInternational conference on machine learning,...

2023
[43]

Dinov2: Learning robust visual features without supervision.arXiv preprint arXiv:2304.07193, 2023

Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy V o, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, et al. Dinov2: Learning robust visual features without supervision.arXiv preprint arXiv:2304.07193, 2023

2023 arXiv
[44]

Umap: Uniform manifold approximation and projection for dimension reduction.arXiv preprint arXiv:1802.03426, 2018

Leland McInnes, John Healy, and James Melville. Umap: Uniform manifold approximation and projection for dimension reduction.arXiv preprint arXiv:1802.03426, 2018

2018 arXiv
[45]

Least squares quantization in pcm.IEEE transactions on information theory, 28(2):129–137, 1982

Stuart Lloyd. Least squares quantization in pcm.IEEE transactions on information theory, 28(2):129–137, 1982

1982
[46]

A mathematical theory of communication.The Bell system technical journal, 27(3): 379–423, 1948

Claude E Shannon. A mathematical theory of communication.The Bell system technical journal, 27(3): 379–423, 1948

1948
[47]

Decoupled weight decay regularization.arXiv preprint arXiv:1711.05101, 2017

Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization.arXiv preprint arXiv:1711.05101, 2017

2017 arXiv
[48]

Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2014

Diederik P Kingma. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2014

2014 arXiv
[49]

Focal loss for dense object detection

Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Dollár. Focal loss for dense object detection. InProceedings of the IEEE international conference on computer vision, pages 2980–2988, 2017

2017
[50]

Optimal whitening and decorrelation.The American Statistician, 72(4):309–314, 2018

Agnan Kessy, Alex Lewin, and Korbinian Strimmer. Optimal whitening and decorrelation.The American Statistician, 72(4):309–314, 2018. 20

2018
[2015]

doi: 10.1007/s11263-015-0816-y

doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Cell painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes.Nature protocols, 11(9):1757–1774, 2016

Mark-Anthony Bray, Shantanu Singh, Han Han, Chadwick T Davis, Blake Borgeson, Cathy Hartland, Maria Kost-Alimova, Sigrun M Gustafsdottir, Christopher C Gibson, and Anne E Carpenter. Cell painting, a high-content image-based assay for morphological profiling using multiplexed fluorescent dyes.Nature protocols, 11(9):1757–1774, 2016

work page 2016

[2] [2]

Cell painting predicts impact of lung cancer variants.Molecular biology of the cell, 33, 2022

Juan C Caicedo, John Arevalo, Federica Piccioni, Mark-Anthony Bray, Cathy L Hartland, Xiaoyun Wu, Angela N Brooks, Alice H Berger, Jesse S Boehm, Anne E Carpenter, et al. Cell painting predicts impact of lung cancer variants.Molecular biology of the cell, 33, 2022

work page 2022

[3] [3]

Single-cell subcellular protein localisation using novel ensembles of diverse deep architectures

Syed Sameed Husain, Eng-Jon Ong, Dmitry Minskiy, Mikel Bober-Irizar, Amaia Irizar, and Miroslaw Bober. Single-cell subcellular protein localisation using novel ensembles of diverse deep architectures. Communications biology, 6(1):489, 2023

work page 2023

[4] [4]

Unbiased single-cell morphology with self-supervised vision transformers.bioRxiv, 2023

Michael Doron, Théo Moutakanni, Zitong S Chen, Nikita Moshkov, Mathilde Caron, Hugo Touvron, Piotr Bojanowski, Wolfgang M Pernice, and Juan C Caicedo. Unbiased single-cell morphology with self-supervised vision transformers.bioRxiv, 2023

work page 2023

[5] [5]

Subcell: Vision foundation models for microscopy capture single-cell biology.bioRxiv, pages 2024–12, 2024

Ankit Gupta, Zoe Wefers, Konstantin Kahnert, Jan N Hansen, Will Leineweber, Anthony Cesnik, Dan Lu, Ulrika Axelsson, Frederic Ballllosera Navarro, Theofanis Karaletsos, et al. Subcell: Vision foundation models for microscopy capture single-cell biology.bioRxiv, pages 2024–12, 2024

work page 2024

[6] [6]

Self-supervision advances morphological profiling by unlocking powerful image representations.Scientific Reports, 15(1):4876, 2025

Vladislav Kim, Nikolaos Adaloglou, Marc Osterland, Flavio M Morelli, Marah Halawa, Tim König, David Gnutt, and Paula A Marin Zapata. Self-supervision advances morphological profiling by unlocking powerful image representations.Scientific Reports, 15(1):4876, 2025

work page 2025

[7] [7]

Berg, and Li Fei-Fei

Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, Alexander C. Berg, and Li Fei-Fei. ImageNet Large Scale Visual Recognition Challenge.International Journal of Computer Vision (IJCV), 115(3):211–252,

[8] [8]

A subcellular map of the human proteome

Peter J Thul, Lovisa Åkesson, Mikaela Wiking, Diana Mahdessian, Aikaterini Geladaki, Hammou Ait Blal, Tove Alm, Anna Asplund, Lars Björk, Lisa M Breckels, et al. A subcellular map of the human proteome. Science, 356(6340):eaal3321, 2017

work page 2017