Wei Li

pertTF: context-aware AI modeling for genome-scale and cross-system perturbation prediction

The paper introduces pertTF, a transformer-based model for predicting genetic perturbation responses, showing improved performance in expression prediction and cell identity shifts across cellular contexts.

Perturb-seq reveals distinct responses to pluripotency regulator dosages underlying the control of self-renewal and differentiation

This study uses Perturb-seq to analyze how varying levels of pluripotency regulators affect hESC self-renewal and differentiation through single-cell RNA-seq.

Targeting the translation initiation complex in therapy-resistant and metastatic melanoma

This study identifies M19-6 as a small molecule that disrupts the eIF4F complex, reducing melanoma growth and metastasis while sparing melanocytes.

Decoding Heterogenous Single-cell Perturbation Responses

The paper introduces a method to quantify diverse perturbation responses and discover novel biological insights in single-cell perturbation datasets.

Deep Domain Adversarial Neural Network for the Deconvolution of Cell Type Mixtures in Tissue Proteome Profiling

This paper introduces scpDeconv, a deep learning method for deconvolving cell type mixtures in proteomic data, showing improved performance across species and technologies.

Modeling CRISPR-Cas13d on-target and off-target effects using machine learning approaches

The paper develops DeepCas13, a deep learning model to predict CRISPR-Cas13d gRNA on-target activity and off-target viability effects, improving accuracy for RNA editing applications.

A Versatile Deep Graph Contrastive Learning Framework for Single-cell Proteomics Embedding

The paper introduces scPROTEIN, a deep graph contrastive learning framework for single-cell proteomics that addresses uncertainty, denoising, batch effects, and embedding learning.

Intrinsic RNA targeting constrains the utility of CRISPR-Cas13 systems

The paper evaluates CRISPR-Cas13 systems and finds that they have intrinsic RNA targets in mammalian cells, which can affect lentiviral particle production and limit their utility.

Diversification of molecularly defined myenteric neuron classes revealed by single cell RNA-sequencing

The paper uses single-cell RNA-sequencing to define 12 molecularly distinct enteric neuron classes in mice and identifies their developmental trajectories and regulatory mechanisms.

Multi-omics analysis of chromatin accessibility and interactions with transcriptome by HiCAR

The paper introduces HiCAR, a method that simultaneously assesses chromatin accessibility, interactions, and transcriptome in low-input samples, revealing 46,792 cis-regulatory interactions in H1 hESCs.

Linking genotypes with multiple phenotypes in single-cell CRISPR screens

This paper introduces scMAGeCK, a computational framework for analyzing single-cell CRISPR screens to identify genes and non-coding elements associated with multiple phenotypes.

Single cell atlas of trisomy 21 cerebral cortex

This study creates the first single cell atlas of T21 cerebral cortex, identifying dysregulated genes and developing a platform for data exploration.

Quantitative assessment of cell population diversity in single-cell landscapes

This paper introduces sc-UniFrac, a framework to quantify compositional diversity in cell populations across single-cell transcriptome datasets, demonstrating its utility in various applications.

CRISPR/Cas9 screens Reveal Dasatinib Targets of Inhibiting T cell Activation and Proliferation

This study identifies dasatinib targets in T cell activation and proliferation using CRISPR/Cas9 screening in Jurkat cells, revealing both known and novel targets and their functional roles.

Reducing False Positives in CRISPR/Cas9 Screens from Copy Number Variations

This paper develops methods to reduce false positives in CRISPR/Cas9 screens caused by copy number variations, improving screening accuracy.

Integrative analysis and refined design of CRISPR knockout screens

The paper introduces MAGeCK-NEST, a method to improve CRISPR screen analysis by integrating PPI data and refining sgRNA design, leading to better hit calling and library design.