perturbation screen

A 100-million-cell atlas: 50 cancer lines pooled as spheroid villages, treated with 379 drugs at several doses, profiled by combinatorial-barcoding scRNA-seq and embedded with a 10-dimensional scVI model.

This paper introduces a consensus clustering method to identify robust signals from high-throughput screens by analyzing multi-scale data organization in cancer cell lines and bacterium models.

This paper introduces PeML, a similarity metric learning method for perturbational datasets, improving the identification of perturbation mechanisms and biological relationships.

Consensus likelihoods from general, unsupervised protein language models evolve seven human antibodies using only a wildtype sequence and <=20 tested variants, improving binding affinity up to 160-fold.

The paper reformulates modular response analysis as a multilinear regression problem to improve stability and performance in inferring biological networks from perturbation data.

Cell Painting images and profiles for 115,796 compounds in U2OS cells from 12 laboratories; 15.2% of analyzed compounds are phenotypically active, and cross-laboratory replicates cut reference-compound activity by 34%.

The paper introduces an automated system called Lila to design and optimize microbial strains for producing small molecules, achieving high titers for naringenin and creating strains for 242 molecules.

DepLink links genetic and pharmacologic dependencies in cancer by integrating CRISPR, pharmacologic, and expression data through an R Shiny app.

The paper identifies Hnrnpk as a key regulator of prion propagation through a genome-wide RNAi screen using the QUIPPER assay, showing PSA's ability to reduce prion levels and toxicity in multiple models.

This paper introduces structured hierarchical models for probabilistic inference of causal effects from perturbation screening data, showing that incorporating prior knowledge improves parameter inference in noisy datasets.

MultiCPA predicts multimodal perturbation responses at the single-cell level using a generative model trained on CRISPR perturbation data.

The paper develops a model to infer pairwise epistatic gene interactions from siRNA perturbation screens, showing that glinternet outperforms xyz in accuracy for high-dimensional data.

The paper introduces a hybrid machine learning approach for perturbation biology to identify combinatorial therapeutic candidates using a melanoma cell line dataset.

This paper shows ozonide antimalarials disrupt Plasmodium falciparum haemoglobin digestion, leading to parasite death and stress responses.

The paper identifies MTHFD1 as a target for broad antiviral therapy by screening bat cells with RNAi and CRISPR libraries against mumps and influenza viruses.

The paper improves the SATAY method using homologous recombination to enhance transposition efficiency, enabling faster and more comprehensive functional genomics screens in yeast.

This paper introduces a network inference method based on response logic that outperforms existing approaches in benchmark challenges and generates plausible network hypotheses from perturbation data.

The paper trains a random forest classifier, ForestMHC, to predict MHC-I peptide presentation using a million human peptides, outperforming existing tools on test sets and mass spectrometry data.

The paper introduces a transposon-based system for genetic manipulation of the vertebrate CNS, enabling long-term conditional experiments across cell types and species.

MOFA, a sparse Bayesian factor model with mixed likelihoods, separates shared from modality-specific multi-omic variation, recovering IGHV, trisomy 12 and oxidative-stress axes in 200 leukaemia samples.

The paper links transcriptomic, proteomic, and peptidomic brain data to their original brain regions using digital atlas mapping techniques, focusing on the hypothalamus.

In suspended human keratinocytes, five induced phosphatases dampen ERK to drive differentiation commitment, opposed by DUSP10, in a time-rewired network that renders commitment an unstable switch.