Marinka Zitnik - Associate Professor

Marinka Zitnik

Associate Professor

Boston, Massachusetts, United States

Join Prog.AI to see contacts

Summary

🤩

Rockstar

🎓

Top School

Marinka Zitnik is an Associate Professor and computational biologist with 14 years of experience at the intersection of machine learning, network biology, and software engineering, currently based at Harvard and affiliated with the Broad Institute. She builds scalable graph-based ML models for multirelational link prediction and has contributed core algorithmic and engineering improvements to open-source projects—examples include implementing NNDSVD initialization for NMF and optimizing the Decagon graph convolutional model. Her background spans rigorous academic training (PhD and postdoctoral work at institutions including Stanford and University of Ljubljana) and hands-on software development, from SVM integration in Orange3 to performance and compatibility fixes across Python ecosystems. Known for translating complex mathematical methods into robust, well-documented code, she combines research leadership with practical tooling that enables reproducible computational biology.

14 years of coding experience

5 years of employment as a software developer

PhD Student Bioinformatics, PhD Student Bioinformatics at Imperial College London

PhD Student Bioinformatics, PhD Student Bioinformatics at University of Toronto

Predoctoral Fellow Bioinformatics, Predoctoral Fellow Bioinformatics at Baylor College of Medicine

Doctor of Philosophy (Ph.D.) Computer Science, Doctor of Philosophy (Ph.D.) Computer Science at University of Ljubljana

Postdoc Computer Science, Postdoc Computer Science at Stanford University

Slovenian, English, German

Stackoverflow

Stats

1reputation

0reached

0answers

0questions

Github Skills (24)

scipy10

sparse-array10

python10

classification10

scikit10

data-mining10

sparse-matrix10

machine-learning10

svm10

sparse-file10

numpy10

matrix-factorization10

deep-learning10

tensorflow10

scikit-learn10

Programming languages (4)

CSSHTMLJupyter NotebookPython

Github contributions (5)

mims-harvard/decagon

Jan 2018 - Sep 2018

Graph convolutional neural network for multirelational link prediction

Role in this project:

ML Engineer

Contributions:11 commits, 27 pushes, 1 branch in 7 months

Contributions summary:Marinka focused on developing and refining the Decagon model, a graph convolutional neural network for multirelational link prediction. Their contributions include implementing a minibatch iterator for efficient data loading and training, fixing bugs related to model testing, and optimizing the code for improved performance. The user also made changes to support Python 2 and 3 compatibility and represent undirected networks.

graph-convolutional-networkspredictiondeep-learningembeddingsrepresentation-learning

mims-harvard/nimfa

Jul 2011 - Feb 2020

Nimfa: Nonnegative matrix factorization in Python

Role in this project:

Back-end Developer

Contributions:9 releases, 527 commits, 6 PRs in 8 years 8 months

Contributions summary:Marinka focused on developing the core framework of the project with Python. Their contributions included the implementation of a matrix factorization framework that could combine seeding methods and algorithms and also provided a class for storing NMF results. They added a new initialization approach, Nonnegative Double Singular Value Decomposition (NNDSVD), and made various documentation changes to better describe the project. Additionally, the user implemented mathematical formulas and conversions to provide better accuracy on sparse matrices.

pythonfactorizationembeddingsmatrix-factorizationlatent-variable-models

Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.

Request Free Trial