Seyednami Niyakan

Seyednami Niyakan

Ph.D. candidate

Texas A&M University

Biography

I'm a PhD candidate in Genomic Signal Processing (GSP) lab in Electrical and Computer Engineering department of Texas A&M University working under supervision of professor Xiaoning Qian. My research interests are Bayesian Machine Learning and its application in analyzing heterogeneous and high dimensional biological data with a focus on omics data analysis.

A significant part of my research is related to Bayesian deep learning models and their application in analyzing high dimensional data such as omics data. With this goal in mind, I have developed and published several Bayesian models and computational tools for analyzing different omics data technologies such as:

  • EXPORT: VAE-based model for ordinally purturbed transcriptomics data analysis [Paper]
  • MUSTANG: Spatial Transcriptomics data analysis tool [Github] [Paper]
  • SimCD: Single-cell RNA-seq data analysis tool [Github] [Paper]
  • Pathway-based analyses of radiated bulk RNA-seq data [Paper]
  • Joint Metabolomics and RNA-seq transcriptomics data analysis pipeline [Paper]

    • As of having interest in solving challenges in scRNA-seq data analysis, I participated in IEEE COVID-19 Single-cell transcriptomics Data Hackathon where I won first place by presenting Biomarker identification for COVID-19 severity based on BALF scRNA-seq data. To learn more about this work you can check on this Github page.

    Apart from academic research, I have experience working in collaborative industry environments right alongside engineers, scientists and biologists. During summer of 2021 I did an internship as a Bioinformatics scientist with Department of Quantitative health sciences at Mayo Clinic. My internship project was mainly focused on gene-drug relationship discovery and learning interactions between genes, drugs and diseases. In summer of 2022, I joined Genentech as a data scientist intern and I worked on developing pipelines for integrating and analyzing large single-cell RNA-seq datasets.

    Interests

    • Bayesian Deep Learning
    • Machine Learning
    • Multi-Modal Data Analysis
    • Computational Biology
    • Multi-omics Data Analysis

    Education

    • Ph.D. in Electrical & Computer Engineering, 2019 - present

      Texas A&M University

    • M.Sc. in Electrical & Computer Engineering, 2018 - 2021

      Texas A&M University

    • B.Sc. in Electrical Engineering, 2013 - 2018

      Sharif University of Technology

    News

    Featured Publications

    Biologically Interpretable VAE with Supervision for Transcriptomics Data Under Ordinal Perturbations

    Latent variable models such as the Variational Auto-Encoders (VAEs) have shown impressive performance for inferring expression patterns for cell subtyping and biomarker identification from transcriptomics data. However, the limited interpretability of their latent variables obscures deriving meaningful biological understanding of cellular responses to different external and internal perturbations. We here propose a novel deep learning framework, EXPORT (EXPlainable VAE for ORdinally perturbed Transcriptomics data), for analyzing ordinally perturbed transcriptomics data that can incorporate any biological pathway knowledge in the VAE latent space. With the corresponding pathway-informed decoder, the learned latent expression patterns can be explained as pathway-level responses to perturbations, offering direct interpretability with biological understanding.
    S. Niyakan, X. Luo, BJ. Yoon and X. Qian
    PDF
    EXPORT

    MUSTANG: MUlti-sample Spatial Transcriptomics data ANalysis tool with cross-sample transcriptional Guidance

    Spatially resolved transcriptomics has revolutionized genome-scale transcriptomic profiling by providing high-resolution characterization of transcriptional patterns. We here present our spatial transcriptomics analysis framework, MUSTANG (MUlti-sample Spatial Transcriptomics data ANalysis with cross-sample transcriptional similarity Guidance), which is capable of performing multi-sample spatial transcriptomics spot cellular deconvolution by allowing both cross-sample expression based similarity information sharing as well as spatial correlation in gene expression patterns within samples.
    S. Niyakan, J. Sheng, Y. Cao, et al.
    PDF Code
    MUSTANG

    SimCD: Simultaneous Clustering and Differential expression analysis for single-cell transcriptomic data

    Single-Cell RNA sequencing (scRNA-seq) measurements have facilitated genome-scale transcriptomic profiling of individual cells, with the hope of deconvolving cellular dynamic changes in corresponding cell sub-populations to better understand molecular mechanisms of different development processes...
    S. Niyakan, E. Hajiramezanali, Sh. Boluki, et al.
    PDF Code
    SimCD

    Biomarker identification for COVID-19 severity based on BALF scRNA-seq data

    Recently, Single-cell RNA sequencing (scRNA-seq) transcriptome profiling of Bronchoalveolar lavage fluid (BALF) cells has been done to elucidate the potential mechanisms underlying in COVID-19. With the aim of better utilizing this atlas of BALF cells in response to the virus,...
    S. Niyakan, X. Qian
    PDF Code Slides
    Biomarker identification for COVID-19 severity based on BALF scRNA-seq data