JUMP Cell Painting dataset: morphological impact of 136,000 chemical and genetic perturbations

Chandrasekaran, Srinivas Niranj; Ackerman, Jeanelle; Alix, Eric; Ando, D. Michael; Arevalo, John; Bennion, Melissa; Boisseau, Nicolas; Borowa, Adriana; Boyd, Justin D.; Brino, Laurent; Byrne, P.J.; Ceulemans, Hugo; Ch’ng, Carolyn; Cimini, Beth A.; Clevert, Djork-Arné; Deflaux, Nicole; Doench, John G.; Doyonnas, Régis; Dragone, Vincenza; Engkvist, Ola; Faloon, Patrick W.; Fritchman, Briana; Fuchs, Florian S.; Garg, Sakshi; Gilbert, Tamara J.; Glazer, David; Gnutt, David; Goodale, Amy; Grignard, Jeremy; Guenther, Judith; Han, Yu; Hanifehlou, Zahra; Hariharan, Santosh; Hernandez, Desiree; Horman, Shane R.; Hormel, Gisela; Huntley, Michael; Icke, Ilknur; Iida, Makiyo; Gros, Christina; Jaensch, Steffen; Khetan, Jawahar; Kost‐Alimova, Maria; Krawiec, Tomasz; Kühn, Daniel; Lardeau, Charles-Hugues; Lembke, Amanda; Lin, Francis; Little, Kevin D.; Lofstrom, Kenneth R.; Lofti, Sofia; Logan, David J.; Luo, Yi; Madoux, Franck; Zapata, Paula Andrea Marin; Marion, Brittany A.; Martin, Glynn; McCarthy, Nicola; Mervin, Lewis; Miller, Lisa; Mohamed, Haseeb; Monteverde, Tiziana; Mouchet, Elizabeth; Nicke, Barbara; Ogier, Arnaud; Ong, Anne-Laure; Osterland, Marc; Otrocka, Magdalena; Peeters, Pieter J.; Pilling, James; Prechtl, Stefan; Chen, Qian; Rataj, Krzysztof; Root, David E.; Sakata, Sylvie K.; Scrace, Simon; Shimizu, Hajime; Simón, David; Sommer, Peter; Spruiell, Craig; Sumia, Iffat; Swalley, Susanne E.; Terauchi, Hiroki; Thibaudeau, Amandine; Unruh, Amy; Waeter, Jelle Van de; Dyck, Michiel Van; Staden, Carlo van; Warchoł, Michał; Weisbart, Erin; Weiss, Amélie; Wiest-Daesslé, Nicolas; Williams, Guy; Yu, Shan; Zapiec, Bolek; Zyla, Marek; Singh, Shantanu; Carpenter, Anne E.

doi:10.1101/2023.03.23.534023

Cited by 33 publications

(59 citation statements)

References 32 publications

(68 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5c, DINO consistently outperformed CellProfiler in classifying gene families for a wide range of k nearest neighbors in terms of F1-score. In general, the performance metrics were much lower for ORF perturbations than for chemical perturbations, consistent with the previously reported lower reproducibility of ORFs over compounds 35 .…”

Section: Hierarchical Clustering Of Dino Representations Recapitulate...supporting

confidence: 89%

“…In contrast to previous works that only focused on smaller scale Cell Painting datasets (BBBC022 53 , BBBC036 54 , BBBC037 55 , BBBC047 54 ), we are the first to present a holistic selfsupervised study that leverages the JUMP-CP consortium data 35 . Still, we are only using a subset of the full JUMP-CP dataset.…”

Section: Discussionmentioning

confidence: 99%

“…Each compound in the JUMP-CP dataset 35 was screened in 5 replicates across 12 laboratories ("sources") using a standardized protocol but possibly different instrumentation, which allowed us to study the scaling properties of SSL models in relation to training data size and heterogeneity. We used two training sets (Fig.…”

Section: Jump-cp Datasets For Benchmarking Ssl Representationsmentioning

confidence: 99%

“…1c). The single-source and multisource evaluation sets were constructed using JUMP Target2 plates (see Methods and 35 ) from the respective sources of the single-source and multisource training sets. JUMP Target2 plates contain compounds with target annotations and were imaged in every experimental batch, which enabled us to not only assess model performance using biological labels but also evaluate batch and laboratory effects.…”

Section: Jump-cp Datasets For Benchmarking Ssl Representationsmentioning

confidence: 99%

“…The JUMP-CP consortium recently released one of the largest public Cell Painting datasets 35 . This large-scale image set (200 TB) with approximately 116,000 compounds and 13,000 genetic perturbations (CRISPR and ORF) offers unique opportunities for developing novel AI-based morphological profiling methods.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Self-supervision advances morphological profiling by unlocking powerful image representations

Kim

Adaloglou

Osterland

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Morphological profiling is a powerful technology that enables unbiased characterization of cellular states through image-based screening. Inspired by recent progress in self-supervised learning (SSL), we sought to explore the potential benefits of using SSL in this domain and conducted a comprehensive benchmark study of recent SSL methods for learning representations from Cell Painting images without segmentation. We trained DINO, MAE, and SimCLR on subsets of the JUMP-CP consortium data, one of the largest publicly available Cell Painting image sets, and observed improved model performance with larger and more heterogeneous training sets. Our best model (DINO) surpassed the widely used profiling tool CellProfiler by 29% in mean average precision (mAP) on classifying chemical perturbations and significantly accelerated feature extraction by 50x, at a lower cost. Moreover, DINO outperformed CellProfiler in clustering gene families on an independent gene overexpression dataset. Our findings indicate that SSL methods can improve the efficiency and performance of morphological profiling, offering the potential to expedite drug discovery and reduce compute costs.

show abstract

Section: Hierarchical Clustering Of Dino Representations Recapitulate...supporting

confidence: 89%

Section: Discussionmentioning

confidence: 99%

Section: Jump-cp Datasets For Benchmarking Ssl Representationsmentioning

confidence: 99%

Section: Jump-cp Datasets For Benchmarking Ssl Representationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Self-supervision advances morphological profiling by unlocking powerful image representations

Kim

Adaloglou

Osterland

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Assessment of Drug-Induced Liver Injury through Cell Morphology and Gene Expression Analysis

Lejal,

Cerisier,

Rouquié

et al. 2023

Chem. Res. Toxicol.

View full text Add to dashboard Cite

Drug-induced liver injury (DILI) is a significant concern in drug development, often leading to drug withdrawal. Although many studies aim to identify biomarkers and gene/pathway signatures related to liver toxicity and aim to predict DILI compounds, this remains a challenge in drug discovery. With a strong development of high-content screening/imaging (HCS/HCI) for phenotypic screening, we explored the morphological cell perturbations induced by DILI compounds. In the first step, cell morphological signatures were associated with two datasets of DILI chemicals (DILIRank and eTox). The mechanisms of action were then analyzed for chemicals having transcriptomics data and sharing similar morphological perturbations. Signaling pathways associated with liver toxicity (cell cycle, cell growth, apoptosis, ...) were then captured, and a hypothetical relation between cell morphological perturbations and gene deregulation was illustrated within our analysis. Finally, using the cell morphological signatures, machine learning approaches were developed to predict chemicals with a potential risk of DILI. Some models showed relevant performance with validation set balanced accuracies between 0.645 and 0.739. Overall, our findings demonstrate the utility of combining HCI with transcriptomics data to identify the morphological and gene expression signatures related to DILI chemicals. Moreover, our protocol could be extended to other toxicity end points, offering a promising avenue for comprehensive toxicity assessment in drug discovery.

show abstract

Insights into Drug Cardiotoxicity from Biological and Chemical Data: The First Public Classifiers for FDA Drug-Induced Cardiotoxicity Rank

Seal,

Spjuth,

Hosseini-Gerami

et al. 2024

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Drug-induced cardiotoxicity (DICT) is a major concern in drug development, accounting for 10−14% of postmarket withdrawals. In this study, we explored the capabilities of chemical and biological data to predict cardiotoxicity, using the recently released DICTrank data set from the United States FDA. We found that such data, including protein targets, especially those related to ion channels (e.g., hERG), physicochemical properties (e.g., electrotopological state), and peak concentration in plasma offer strong predictive ability for DICT. Compounds annotated with mechanisms of action such as cyclooxygenase inhibition could distinguish between mostconcern and no-concern DICT. Cell Painting features for ER stress discerned most-concern cardiotoxic from nontoxic compounds. Models based on physicochemical properties provided substantial predictive accuracy (AUCPR = 0.93). With the availability of omics data in the future, using biological data promises enhanced predictability and deeper mechanistic insights, paving the way for safer drug development. All models from this study are available at https://broad.io/DICTrank_Predictor.

show abstract

JUMP Cell Painting dataset: morphological impact of 136,000 chemical and genetic perturbations

Cited by 33 publications

References 32 publications

Self-supervision advances morphological profiling by unlocking powerful image representations

Self-supervision advances morphological profiling by unlocking powerful image representations

Assessment of Drug-Induced Liver Injury through Cell Morphology and Gene Expression Analysis

Insights into Drug Cardiotoxicity from Biological and Chemical Data: The First Public Classifiers for FDA Drug-Induced Cardiotoxicity Rank

Contact Info

Product

Resources

About