Unifying Genetic and Chemical Perturbagen Representation through a Hybrid Deep Learning Framework DOI Open Access
Yiming Li, Jun Zhu, Linjing Liu

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 7, 2025

The integration of genetic and chemical perturbations has driven transformative advances in elucidating cellular mechanisms accelerating drug discovery. However, the lack a unified representation for diverse perturbagen types limits comprehensive analysis joint modeling multi-domain perturbation agents (molecular cause space) their resulting phenotypes (phenotypic effect spaces). Here, we present UniPert, hybrid deep learning framework that encodes perturbagens into shared semantic space. UniPert employs tailored encoders to address inherent molecular-scale differences across leverages contrastive with experiment-driven compound-target interactions bridge these domains. Extensive experiments validate UniPert’s versatility application. generated representations effectively capture hierarchical pharmacological relationships perturbagens, facilitating annotations understudied targets compounds. can be plugged advanced frameworks enhance performance both outcome prediction tasks. Notably, paves way cross-domain modeling, driving novel genetic-to-chemical transfer paradigm, boosting context-specific silico screening efficiency development personalized therapies.

Language: Английский

Spatially Resolved in vivo CRISPR Screen Sequencing via Perturb-DBiT DOI Creative Commons
Alev Baysoy, Xiaolong Tian, Feifei Zhang

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 19, 2024

Perturb-seq enabled the profiling of transcriptional effects genetic perturbations in single cells but lacks ability to examine impact on tissue environments. We present Perturb-DBiT for simultaneous co- sequencing spatial transcriptome and guide RNAs (gRNAs) same section vivo CRISPR screen with genome-scale gRNA libraries, offering a comprehensive understanding how modifications affect cellular behavior architecture. This platform supports variety delivery vectors, library sizes, preparations, along two distinct capture methods, making it adaptable wide range experimental setups. In applying Perturb-DBiT, we conducted un-biased knockouts tens genes or at genome-wide scale across three cancer models. mapped all gRNAs individual colonies corresponding transcriptomes human metastatic colonization model, revealing clonal dynamics cooperation. also examined effect perturbation tumor immune microenvironment an immune-competent syngeneic uncovering differential synergistic promoting infiltration suppression tumors. allows simultaneously evaluating each knockout initiation, development, metastasis, histopathology, landscape. Ultimately, not only broadens scope inquiry, lays groundwork developing targeted therapeutic strategies.

Language: Английский

Citations

8
Unifying Genetic and Chemical Perturbagen Representation through a Hybrid Deep Learning Framework DOI Open Access
Yiming Li, Jun Zhu, Linjing Liu

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 7, 2025

The integration of genetic and chemical perturbations has driven transformative advances in elucidating cellular mechanisms accelerating drug discovery. However, the lack a unified representation for diverse perturbagen types limits comprehensive analysis joint modeling multi-domain perturbation agents (molecular cause space) their resulting phenotypes (phenotypic effect spaces). Here, we present UniPert, hybrid deep learning framework that encodes perturbagens into shared semantic space. UniPert employs tailored encoders to address inherent molecular-scale differences across leverages contrastive with experiment-driven compound-target interactions bridge these domains. Extensive experiments validate UniPert’s versatility application. generated representations effectively capture hierarchical pharmacological relationships perturbagens, facilitating annotations understudied targets compounds. can be plugged advanced frameworks enhance performance both outcome prediction tasks. Notably, paves way cross-domain modeling, driving novel genetic-to-chemical transfer paradigm, boosting context-specific silico screening efficiency development personalized therapies.

Language: Английский

Citations

0