modprofin — Your Data Analysis Tool, From Biology to Markets

Capabilities

Everything in one
native desktop app

No cloud. No subscription. Runs entirely on your machine.

🔬

Bioinformatics

Protein Function Prediction

CAFA-6 pipeline with k-mer similarity and TDA + Random Forest. Predicts Gene Ontology terms across MF, BP, and CC ontologies for any amino acid sequence.

Single protein & batch mode
K-mer similarity index
Persistent homology features
GO hierarchy propagation

🏥

Medicine

Medical Diagnosis Engine

58 conditions, 265 symptoms and 27 lab tests. Weighted differential scoring with severity classification and clinical notes — completely offline.

Symptom / lab / risk intake
Differential ranking
Lab critical value detection
GO → disease bridge

🧬

Genomics

DNA Analysis Suite

From raw nucleotide to clinical insight in seconds. SNP annotation, polygenic risk scores, pharmacogenomics, 6-frame ORF finding and CpG island detection.

75+ curated SNP variants
8-trait PRS models
CYP2D6, CYP2C19, HLA-B alerts
6-frame protein translation

🔮

Quantitative Finance

Nash-Equilibrium Market Predictor

Vietoris-Rips persistent homology with Nash Equilibrium modulation at every filtration stage. Topological regime detection meets game theory.

Nash-filtered distance matrix
H0 / H1 / H2 Betti curves
6 market regime classes
STRONG BUY → STRONG SELL signal

💹

Trading

Live Trading Platform

Paper-trade with real Yahoo Finance prices or connect to Alpaca Markets. Full order management, portfolio tracking, and one-click signal execution.

Paper broker (no account needed)
Alpaca paper & live API
Market / limit / stop orders
"Execute Signal →" one-click

⚖

Risk

Risk Manager

Position sizing, stop-loss / take-profit calculator, risk/reward ratio, max daily loss limits. Know your numbers before every trade.

Kelly & vol-parity sizing
Auto-size from equity %
Real-time P&L tracking
Portfolio allocation view

Inside the app

One platform,
eight specialisations

Protein function prediction at scale

Built on the CAFA-6 dataset with a dual-model architecture. The K-mer Similarity model finds structurally similar proteins in milliseconds. The Full Model combines persistent homology (Vietoris-Rips) with Random Forest for deeper predictions.

Prediction output

GO:0003824  catalytic activity         0.9241
GO:0016740  transferase activity        0.8810
GO:0005488  binding                     0.8503
GO:0044237  cellular metabolic process  0.7991
GO:0009987  cellular process            0.7650

TDA Pipeline

H0 components4

H1 loops7

Betti entropy1.82

Persistence0.443

Personalised differential diagnosis

Enter symptoms, lab values and risk factors. The engine scores 58 conditions using a weighted three-layer model (symptoms 45 %, labs 35 %, risk factors 20 %) and ranks them by probability with severity flags.

Type 2 Diabetes 91%

Metabolic Syndrome 74%

Hypothyroidism 48%

Cushing's Syndrome 31%

Lab Values

HbA1c8.2 %↑ HIGH

Fasting glucose182 mg/dL↑ HIGH

LDL142 mg/dL↑ BORDER

HDL38 mg/dL↓ LOW

TSH3.2 mIU/LNORMAL

eGFR61 mL/min↓ WATCH

Persistent homology meets Nash Equilibrium

At every filtration step ε_k the algorithm solves a Nash Equilibrium for the active asset sub-game and uses the NE strategy alignment to modulate the Vietoris-Rips distance matrix. The resulting topology encodes strategic market structure.

Nash-filtered VR algorithm

for k in range(n_stages):
  ε_k  = k/K · D_max
  # solve Nash equilibrium
  w*   = nash_min_variance(Σ_active)
  conc = max(w*) · N
  # modulate distances
  for (i,j) in active_pairs(ε_k):
    sim      = tanh(r_i @ w* · r_j @ w*)
    D_k[i,j] = D[i,j] · (1 − α·conc·sim)
D_nash = mean(D_k)

Signal Output

STRONG BUY

Confidence84%

RegimeMEAN-REVERSION

H1 loops5

NE concentration0.312

Nash entropy1.94

From signal to execution in one click

The "Execute Signal →" button reads the Nash predictor output, sizes positions using risk parameters, and submits market orders to the paper broker (real Yahoo Finance prices) or directly to Alpaca Markets paper/live API.

🔮

Signal generated

→

⚖

Position sized

→

📤

Order submitted

→

✅

Filled

Portfolio

AAPL20 sh+$142.60

MSFT15 sh+$89.25

GLD30 sh−$23.10

TLT50 sh−$41.00

Total P&L+$167.75

Cash$76,420

Creator

PhD

Dr. Zakaria Lamine

PhD · Applied Mathematics

Zakaria Lamine is an applied mathematician whose research bridges topological data analysis, game theory, and computational biology. His doctoral work formalised the use of persistent homology for high-dimensional biological sequence analysis, leading to the Nash-filtered Vietoris-Rips pipeline at the core of modprofin.

Frustrated by the fragmentation of bioinformatics, clinical genomics, and quantitative finance tooling across dozens of incompatible platforms, he designed modprofin as a single coherent desktop environment — from protein GO prediction to live order execution — grounded in rigorous mathematics.

Persistent Homology

Nash Equilibrium Theory

Computational Genomics

Quantitative Finance

Machine Learning

Topological Data Analysis

Differential Geometry

Stochastic Processes

zakarialamine2@gmail.com +212 663 35 75 20

13+ Years in mathematics

6 Platform modules

∞ Curiosity

"Mathematics is the language nature uses to keep its secrets — and the only tool precise enough to decode them."

📋 Education & Career

2019 – Present

Research & Development — Topological Methods in Biology & Finance

Independent Research · modprofin

Ongoing research applying persistent homology, game theory, and machine learning to biological sequence analysis and quantitative finance. Developed the Nash-filtered Vietoris-Rips pipeline, polygenic risk score models, and pharmacogenomics annotation frameworks. Culminated in the design and full implementation of the modprofin platform — integrating CAFA-6 protein prediction, medical diagnosis, DNA analysis, and live algorithmic trading into a single offline desktop environment.

PhD

PhD — Applied Mathematics

Doctoral Research

Dissertation: "Persistent Diagrams: A Convenient Tool for Proteins Structures/Functions Analysis and Prediction." Developed original theoretical frameworks at the intersection of algebraic topology and computational biology — establishing persistent homology as a rigorous mathematical instrument for protein structure and function analysis. Solved key modality problems arising in the topological representation of biomolecular data, and provided new theoretical foundations for applied mathematics in the life sciences: cohomological invariants of protein complexes, birth-death persistence for structural feature extraction, and formal convergence proofs for diagram-based predictors.

2016 – 2019

MSc — Bioinformatics / Biomathematics

Graduate Studies

Specialisation at the interface of mathematics and life sciences: mathematical modelling of biological systems, sequence analysis, structural bioinformatics, stochastic differential equations in population dynamics, and topological methods for genomic data. Master's thesis on persistent homology applied to protein interaction networks. Graduated with distinction.

2011 – 2016

BSc — Applied Mathematics

Undergraduate Studies

Five-year programme covering real and functional analysis, abstract algebra, differential geometry, probability theory, numerical methods, and optimisation. Undergraduate capstone project: Nash equilibria in zero-sum bimatrix games and their combinatorial structure. First-class honours.

📄 Publications

Peer-reviewed Journals

A Topological Data Analysis of the Protein Structure

International Journal of Analysis and Applications

A Topological Approach for Analyzing the Protein Structure

Communications in Mathematical Biology and Neuroscience

Artificial Intelligence Model Based on Algebraic Topology for Protein Structure Analysis and Prediction

Contemporary Mathematics · WISERPUB

Mathematical Insights into Protein Architecture: Persistent Homology and Machine Learning Applied to the Flagellar Motor

Networks and Heterogeneous Media · AIMS Press Under Review

Preprints

Persistent Diagrams for Protein Structure Prediction

Research Square · Preprint

Mathematical Modeling of Protein Structures: A Cohomology-Based Approach to the Flagellar Motor

arXiv · Preprint

Books

📘

Mathematical Insights into Protein Architectures — Tome 1

Lamine, Z. Book

📗

Mathematical Insights into Protein Architectures — Tome 2

Lamine, Z. Book

Thesis & Academic Reports

✦

Persistent Homology in Detection of Protein Features

MSc Thesis · Supervisor: Prof. MyIsmail Mamouni

✦

Other Academic Reports & Student Projects

Approximation of APPELL Hypergeometric Function via Padé Method (2015) · Hidden Markov Models in R · Solution of Linear Systems in Python

⚡ Technical Skills

Topological Data AnalysisExpert

Python (numpy · scipy · sklearn)Expert

Nash Equilibrium & Game TheoryExpert

Computational Genomics & BioinformaticsAdvanced

Quantitative Finance & Risk ModellingAdvanced

Machine Learning (RF · SVM · DL)Advanced

Differential Geometry & Manifold TheoryAdvanced

Desktop App Development (tkinter · PyInstaller)Proficient

French

English

Arabic

Pricing

One purchase.
Yours forever.

No subscriptions. No recurring fees. Pay once, own the licence permanently.

Monthly Annual Save 40%

🔬

Essential

Life Sciences suite

$ 23 /mo

Billed $276/year

Buy Essential

Protein Function Prediction (CAFA-6)
Single & batch analysis
Medical Diagnosis Engine
58 conditions · 265 symptoms
DNA Analysis Suite
75+ curated SNP variants
Protein-Disease Link tab
Offline · no data leaves device
Market Predictor
Trading Platform
Risk Manager
Alpaca broker integration

YOUR DATA
ANALYSIS TOOL,
FROM BIOLOGY
TO MARKETS

Everything in one
native desktop app

Protein Function Prediction

Medical Diagnosis Engine

DNA Analysis Suite

Nash-Equilibrium Market Predictor

Live Trading Platform

Risk Manager

One platform,
eight specialisations

Protein function prediction at scale

Personalised differential diagnosis

Persistent homology meets Nash Equilibrium

From signal to execution in one click

How the pipeline works

Input

Feature Extraction

Nash-TDA Inference

Results + Action

Open-source
at its core

Dr. Zakaria Lamine

📋 Education & Career

📄 Publications

⚡ Technical Skills

One purchase.
Yours forever.

Already have
a licence?

YOUR DATAANALYSIS TOOL, FROM BIOLOGYTO MARKETS

Everything in onenative desktop app

Protein Function Prediction

Medical Diagnosis Engine

DNA Analysis Suite

Nash-Equilibrium Market Predictor

Live Trading Platform

Risk Manager

One platform,eight specialisations

Protein function prediction at scale

Personalised differential diagnosis

Persistent homology meets Nash Equilibrium

From signal to execution in one click

How the pipeline works

Input

Feature Extraction

Nash-TDA Inference

Results + Action

Open-sourceat its core

Dr. Zakaria Lamine

📋 Education & Career

📄 Publications

⚡ Technical Skills

One purchase.Yours forever.

Already havea licence?

YOUR DATA
ANALYSIS TOOL,
FROM BIOLOGY
TO MARKETS

Everything in one
native desktop app

One platform,
eight specialisations

Open-source
at its core

One purchase.
Yours forever.

Already have
a licence?