Qiao Liu, PhD

bayesgm is a unified framework for Bayesian generative modeling that combines deep neural networks with principled probabilistic inference. It provides tools for learning latent representations, causal inference, conditional inference with quantifying uncertainty under complex settings. It provides both Python and R packages.

Faculty: Qiao Liu, PhD

Download: Liu Lab / bayesgm package

Platform: Python

Reference: arxiv.org (bayesgm)

---

CausalBGM is an innovative Bayesian generative modeling framework tailored for causal inference in observational studies with high-dimensional covariates and large-scale datasets. It addresses key challenges in the field by leveraging AI-powered techniques and Bayesian principles to estimate causal effects, especially individual treatment effects (ITEs), with robust uncertainty quantification.

The novelties of the model include:

1. An Gibbs-type iterative updating algorithm that calculates likelihood on mini-batches in each step, ensuring computational efficiency while accommodating massive datasets.
2. An Encoding Generative Modeling (EGM) initialization strategy to stabilize model training and enhance predictive performance.
3. A fully Bayesian approach that treats model parameters as random variables and models both mean and variance functions, supporting the construction of well-calibrated posterior intervals of causal effect estimates

CausalBGM stands as a scalable, rigorous, and interpretable solution for modern causal inference applications, effectively bridging AI and Bayesian causal inference.

Faculty: Qiao Liu, PhD

Download: Liu Lab / CausalBGM package

Platform: Python

Reference: arxiv.org (CausalBGM)

---

CausalEGM is a generative modeling framework for causal inference that integrates implicit generative models with structured causal representations. It enables estimation of causal effects from observational data by learning latent variables and modeling complex, nonlinear relationships. It provides both Python and R packages.

Faculty: Qiao Liu, PhD

Download: Liu Lab / CausalEGM package

Platform: Python

Reference: pnas.org (CausalEGM)

---

In this study, we present DeepCDR which integrates multi-omics profiles of cancer cells and explores intrinsic chemical structures of drugs for predicting CDR. Specifically, DeepCDR is a hybrid graph convolutional network consisting of a uniform graph convolutional network and multiple subnetworks. Unlike prior studies modeling hand-crafted features of drugs, DeepCDR automatically learns the latent representation of topological structures among atoms and bonds of drugs. Extensive experiments showed that DeepCDR outperformed state-of-the-art methods in both classification and regression settings under various data settings. We also evaluated the contribution of different types of omics profiles for assessing drug response. Furthermore, we provided an exploratory strategy for identifying potential cancer-associated genes concerning specific cancer types. Our results highlighted the predictive power of DeepCDR and its potential translational value in guiding disease-specific drug design.

Faculty: Qiao Liu, PhD

Download: Liu Lab / DeepCDR package

Platform: Python

Reference: doi.org (DeepCDR)

---

EpiGePT is a transformer-based model for cross-cell-line prediction of chromatin states by taking long DNA sequence and transcription factor profile as inputs. This is a script for reproducing EpiGePT using TensorFlow.

Faculty: Qiao Liu, PhD

Download: Liu Lab / EpiGePT package

Platform: Python

Reference: doi.org (EpiGePT)

---

Hi-C is a genome-wide technology for investigating 3D chromatin conformation by measuring physical contacts between pairs of genomic regions. The resolution of Hi-C data directly impacts the effectiveness and accuracy of downstream analysis such as identifying topologically associating domains (TADs) and meaningful chromatin loops. High resolution Hi-C data are valuable resources which implicate the relationship between 3D genome conformation and function, especially linking distal regulatory elements to their target genes. However, high resolution Hi-C data across various tissues and cell types are not always available due to the high sequencing cost. It is therefore indispensable to develop computational approaches for enhancing the resolution of Hi-C data.

Faculty: Qiao Liu, PhD

Download: Liu Lab / hicGAN package

Platform: Python

Reference: doi.org (hicGAN)

---

Roundtrip is a deep generative neural density estimator which exploits the advantage of GANs for generating samples and estimates density by either importance sampling or Laplace approximation. This repository provides source code and instructions for using Roundtrip on both simulation data and real data.

Faculty: Qiao Liu, PhD

Download: Liu Lab / Roundtrip package

Platform: Python

Reference: doi.org (Roundtrip)

---

scDEC is a computational tool for single cell ATAC-seq data analysis with deep generative neural networks. scDEC enables simultaneously learning the deep embedding and clustering of the cells in an unsupervised manner. scDEC is also applicable to multi-modal single cell data. We tested it on the PBMC paired data (scRNA-seq and scATAC-seq) from 10x Genomics (see Tutorials).

Faculty: Qiao Liu, PhD

Download: Liu Lab / scDEC package

Platform: Python

Reference: doi.org (scDEC)

---