Raviteja Vangara

Raviteja Vangara

Postdoctoral Research Scientist

University of California, San Diego

Biography

Dr. Raviteja Vangara, Ph.D. is a Postdoctoral Researcher specializing in the advanced field of machine learning applications for scientific research. He currently holds a position at the esteemed Alexandrov lab within the Department of Cellular and Molecular Medicine at the University of California, San Diego (UCSD). Here, his research focuses on leveraging state-of-the-art machine learning methodologies to conduct insightful mutational signature analyses related to human cancer.

Before joining UCSD, Dr. Vangara contributed his extensive knowledge and innovative approach to research at the Theoretical Division of Los Alamos National Laboratory. His work at LANL covered a broad spectrum of scientific applications, predominantly utilizing unsupervised machine learning techniques such as graphical clustering methods, and non-negative matrix and tensor factorization techniques for advanced pattern recognition and latent feature extraction. During his tenure, Dr. Vangara played a crucial role in the Smart Tensors team, a 2021 R&D award winning collective responsible for releasing several groundbreaking open-source software tools optimized for high-performance computing scientific applications, focusing on scalable distributed computing methods.

Dr. Vangara has been recognized for his academic excellence, earning his Ph.D. with distinction in 2019 for his significant work on Coulombic and non-Coulombic effects of Electric Double Layers. He obtained his M.S. in 2017 from the University of New Mexico’s Petsev lab, Department of Chemical and Biological Engineering, The University of New Mexico.

Download my resumé.

Interests
  • Machine Learning
  • Matrix and Tensor Factorization
  • Cheminformatics and Bioinformatics
Education
  • Ph.D. in Engineering (with Distinction), 2019

    University of New Mexico

  • M.S. in Chemical Engineering, 2017

    University of New Mexico

  • B.Tech. in Chemical Engineering, 2015

    National Institute of Technology, Warangal

Recent News

All news»

May 2024: Geographic variation of mutagenic exposures in kidney cancer genomes paper published in Nature.

Dec 2023: SigProfilerAssignment paper published online on Oxford Bioinformatics.

Aug 2023: SigProfilerMatrixGenerator V2 paper published online on BMC Genomics.

Feb 2023: SigProfilerExtractor paper has been featured in Epidemiology and Genomics Research Program - EGRP’s Research Highlights for 2022.

Nov 2022: Article “Uncovering novel mutational signatures by de novo extraction with SigProfilerExtractor” published online.

Recent Publications

Quickly discover relevant content by filtering publications.
(2023). Assigning mutational signatures to individual samples and individual somatic mutations with SigProfilerAssignment. Bioinformatics.

Cite

(2022). Uncovering novel mutational signatures by de novo extraction with SigProfilerExtractor. Cell Genomics.

Cite

(2021). A neural network for determination of latent dimensionality in non-negative matrix factorization. Machine Learning: Science and Technology.

Cite

(2021). Classical density functional analysis of the ionic size effects on the properties of charge regulating electric double layers. Molecular Physics.

Cite

(2021). Determination of the number of clusters by symmetric non-negative matrix factorization. Big Data III: Learning, Analytics, and Applications.

Cite

(2021). Finding the number of latent topics with semantic non-negative matrix factorization. IEEE Access.

Cite

(2021). Improved protein decoy selection via non-negative matrix factorization. IEEE/ACM Transactions on Computational Biology and Bioinformatics.

Cite

(2021). Nonnegative Tensor-Train Low-Rank Approximations of the Smoluchowski Coagulation Equation. International Conference on Large-Scale Scientific Computing.

Cite

(2021). pydnmfk: Python distributed non negative matrix factorization.

Cite

Projects

.js-id-Cancer-Genetics
Anomalous Diffusion

Anomalous Diffusion

Identification of anomalous diffusion sources by unsupervised learning

Distributed NMFk

Distributed NMFk

Distributed non-negative matrix factorization with determination of the number of latent features

Electrolyte solution structure with surface charge regulation

Electrolyte solution structure with surface charge regulation

Electrolyte solution structure and its effect on the properties of electric double layers with surface charge regulation

Geographic variation of mutagenic exposures in kidney cancer genomes

Geographic variation of mutagenic exposures in kidney cancer genomes

Geographic variation of mutagenic exposures in kidney cancer genomes

Ionic Size effects

Ionic Size effects

Classical density functional analysis of the ionic size effects on the properties of charge regulating electric double layers

Ionic solvation effects

Ionic solvation effects

Ionic solvation and solvent-solvent interaction effects on the charge and potential distributions in electric double layers

Multilayer Perceptron

Multilayer Perceptron

A neural network for the determination of latent dimensionality in non-negative matrix factorization

Multivalent ionic effects

Multivalent ionic effects

Coulombic and non-Coulombic effects in charge-regulating electric double layers

SigProfilerAssignment

SigProfilerAssignment

Assigning mutational signatures to individual samples and individual somatic mutations with SigProfilerAssignment

SigProfilerExtractor

SigProfilerExtractor

Uncovering novel mutational signatures by de novo extraction with SigProfilerExtractor

SigProfilerMatrixGenerator

SigProfilerMatrixGenerator

Visualizing and exploring patterns of large mutational events with SigProfilerMatrixGenerator

Solvophilic and solvophobic surfaces

Solvophilic and solvophobic surfaces

Solvophilic and solvophobic surfaces and non-Coulombic surface interactions in charge regulating electric double layers

SymmNMFk- Graphical clustering with Consensus framework

SymmNMFk- Graphical clustering with Consensus framework

Determination of the number of clusters by symmetric non-negative matrix factorization

The mutagenic forces shaping the genomic landscape of lung cancer in never smokers

The mutagenic forces shaping the genomic landscape of lung cancer in never smokers

The mutagenic forces shaping the genomic landscape of lung cancer in never smokers

Recent & Upcoming Talks

Graph Clustering
Determination of the number of clusters by symmetric non-negative matrix factorization
Topic Modeling
Semantic Nonnegative Matrix Factorization with Automatic Model Determination for Topic Modeling
Topic Modeling

Experience

 
 
 
 
 
Post Doctoral Research Associate
Jan 2021 – Present San Diego
Responsibilities include - Developing and utilizing machine learning approaches for accurate detection of mutational signatures in human cancer.
 
 
 
 
 
Post Doctoral Research Associate
Mar 2020 – Aug 2021 Los Alamos
Responsibilities include:Machine learning developer for applications in a) Epigenetics, specifically breathing dynamics of DNA , b)Chemical Physics and c) Computational Fluid and Solid Mechanics. Specialized in unsupervised machine learning, which involves identification of latent dimensions using Matrix and Tensor Decomposition, Graph-based clustering techniques and Natural Language Processing.
 
 
 
 
 
Graduate Research Assistant
Apr 2018 – Mar 2020 Los Alamos
Responsibilities include:Development of novel matrix and tensor factorization techniques for applications in chemo metrics and phase transitions
 
 
 
 
 
Graduate Research Assistant
Dec 2015 – Mar 2018 Albuquerque, NM
Developing density functional models for charged interfaces involving electrolyte solutions; Built a python framework for classical density functional theory coupled with surface charge regulation; model treats solvent explicitly and accounts surface charge basing on thermodynamic chemical equilibrium. Observed new physical insights and the steps for a molecular theory to address important features like the role of non-columbic interactions - ionic solvation and surface-ion interactions, on the actual electrostatics of the system i.e. the electric double layer.