Loghman Samani

Loghman Samani

Computational Biologist

Diffusion Models • Molecular Dynamics • Machine Learning

About

I hold a master's degree in computational biology from the University of Stuttgart and am currently seeking research or PhD positions that combine computational biology with machine learning. My academic training has focused on building quantitative and principled models for biological systems, with a strong emphasis on methodological clarity and reproducibility.

My research centers on computational structural biology, particularly generative modeling for protein structure design, structure prediction, and molecular dynamics. I am especially interested in physics-aware and geometry-aware generative models, including energy-based and diffusion-based approaches. My goal is to develop models that respect the underlying physical constraints of biomolecular systems and contribute to a deeper, mechanistic understanding of protein folding and conformational dynamics.

Computational Structural Biology Molecular Dynamics Generative AI Protein Modeling Statistical Physics

Projects

CppNet: High-Performance Deep Learning

A C++ deep learning library built for performance and transparency. Features efficient tensor operations via Eigen, CPU parallelism with OpenMP, and GPU acceleration through CUDA for training neural networks at scale.

C++ • Eigen • OpenMP • CUDA

View on GitHub →

TorchDiff: Diffusion Models in PyTorch

PyTorch library for building and experimenting with diffusion-based generative models. Implements DDPM, DDIM, score-based SDE models, latent diffusion, and UnCLIP with support for conditional and unconditional generation.

Python • PyTorch • Diffusion Models

View on GitHub →

GRN-Designer: Spatial Gene Networks

Computational framework for automated design and analysis of gene regulatory networks with spatial structure. Uses hybrid evolutionary and gradient-based optimization to generate networks reproducing predefined spatial expression patterns.

Python • Optimization • PDEs

Master's Thesis Project →

Deep Learning From Scratch

Implementation of core architectures from first principles, including MLPs, CNNs, RNNs with LSTMs, and Transformers. Covers optimization methods and regularization with emphasis on conceptual clarity and mathematical correctness.

Python • Neural Networks • Optimization

View on GitHub →

Biostoch: Stochastic Simulation

Python library for simulating biological systems using deterministic and stochastic formulations. Implements ODE solvers and stochastic reaction models to study cellular and biochemical dynamics under intrinsic noise.

Python • ODEs • Stochastic Simulation

View on GitHub →

SmartSolve: ML Toolkit

Unified framework for implementing and evaluating classical machine learning models. Provides consistent workflow for preprocessing, training, and evaluation across supervised and unsupervised methods.

Python • Machine Learning • Model Evaluation

View on GitHub →

Writing

Latent Diffusion Models and Their Implementation in TorchDiff
Comprehensive guide to latent diffusion models, covering theory, architecture, and practical PyTorch implementation for efficient high-resolution image generation
Medium • 11 min read • Read on Medium →
Score-Based Generative Modeling with SDEs
Detailed exposition of score-based generative models defined via stochastic differential equations, covering VE, VP, and sub-VP formulations with practical PyTorch implementations
Medium • 17 min read • Read on Medium →
TorchDiff: A PyTorch Library for Diffusion Models
Introduction to the theoretical foundations of diffusion models with practical implementations, covering DDPM, DDIM, score-based SDE models, and latent diffusion
Medium • 17 min read • Read on Medium →
From Biology to Computation: Exploring Genetic Algorithms
Step-by-step exploration of genetic algorithms, connecting principles of biological evolution to Python-based implementations for optimization problems
Medium • 18 min read • Read on Medium →
Adam Optimization: Enhancing MLP Performance
Detailed explanation of the Adam optimizer and its application in training multiclass MLPs, highlighting techniques for faster convergence and robust performance
Medium • 21 min read • Read on Medium →
Building a Deep Feedforward Neural Network with Backpropagation
Step-by-step guide to constructing and training deep feedforward networks using backpropagation, with Python implementations for binary classification
Medium • 13 min read • Read on Medium →
Introducing BioStoch: Simulating Biological Systems
Overview of BioStoch, a Python library for deterministic and stochastic simulation of biological systems, including Euler, Runge-Kutta, SSA, Tau-Leaping, and Chemical Langevin methods
Medium • 13 min read • Read on Medium →
Highly Accurate Protein Structure Prediction with AlphaFold
Detailed explanation of AlphaFold, a machine learning system for predicting protein structures from amino acid sequences with atomic accuracy
Medium • 13 min read • Read on Medium →
Introduction to Nuclear Magnetic Resonance (NMR)
Overview of NMR spectroscopy and its applications in structural biology, including protein structure, dynamics, interactions, and intrinsically disordered proteins
Medium • 20 min read • Read on Medium →