Back

Fluid-Squid: DIY Multiplexed Imaging of Cells and Tissues

Hickey, J. W.; Li, H.; Caraccio, C.; Yan, Y.; Marx, K.; Martin, P. M.; Leng, H. T.; Fayiah, J. S.; Towalid, E. F.; Dighero-Kemp, B.; Nolan, G. P.; Prakash, M.; McIlwain, D. R.

2025-10-10 bioengineering
10.1101/2025.10.09.680291 bioRxiv
Show abstract

Recent advances in multiplexed single-cell characterization have revolutionized our insight into cell biology, but many available technologies remain limited by high costs or a lack of customizability. To address these challenges, we developed Fluid-Squid, a cost-effective, quantitative imaging platform that integrates automated fluidics to support customizable, do-it-yourself multiplexed imaging workflows. Using Fluid-Squid, we successfully imaged fresh frozen human intestinal tissues with a 36-plex oligonucleotide-barcoded antibody panel and further demonstrated the feasibility of lyophilizing such multiplexed panels. We also adapted existing multiplexed imaging workflows to characterize individual cells to identify immune cell populations, phenotype, and antigen-specific cells from mouse splenocytes and human peripheral blood mononuclear cells (PBMCs) with a 39-antibody panel. To further expand its utility, we developed a barcoding strategy that allows for the pooling and simultaneous staining of multiple samples, reducing time, costs, and batch effects in single-cell experiments. This approach facilitated rapid titration to optimize antibody concentrations and assess the impact of various blood preparation methods on cell type retention. Overall, our work provides a new open-source framework for automated fluidics and microscopy in a flexible, cost-effective platform, empowering adaptable multiplexed characterization of both single cells and tissues.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.