class ErfanZohrabi:
def __init__(self):
self.role = "MSc Bioinformatics @ University of Bologna"
self.background = "BSc Cellular & Molecular Biology (GPA: 17.87/20)"
self.focus = ["LLMs in Genomics", "Deep Learning for Biology",
"Protein Language Models", "Multi-Omics Integration"]
self.passion = "Decoding life's algorithms through AI & computation"
def current_research(self):
return {
"genomics" : "DNA/Protein sequence classification with Deep Learning",
"llm_bio" : "Large Language Models applied to biological sequences",
"omics" : "Single-cell & multi-omics data integration",
"ai_safety" : "Trustworthy AI in clinical genomics"
}| Domain | Technologies |
|---|---|
| Foundation Models | Transformers · BERT/GPT Architectures · Protein Language Models (PLM) |
| Genomic LLMs | DNABERT · AlphaFold · ESM · ProtTrans |
| Deep Learning | PyTorch · TensorFlow · CNNs · RNNs · Attention Mechanisms |
| Generative AI | VAEs · GANs · Diffusion Models · Sequence Generation |
| Graph Neural Nets | GNNs for PPI Networks · Molecular Graphs (CS224W Stanford) |
| NLP for Biology | Biomedical Text Mining · Drug Discovery · Sequence Tokenization |
┌────────────────────────────────────────────────────────────────────────┐
│ BIOINFORMATICS SKILL TREE │
├─────────────────────┬──────────────────────┬───────────────────────────┤
│ SEQUENCE ANALYSIS │ STRUCTURAL BIO │ OMICS & SYSTEMS │
│ ● DNA Classification│ ● Protein Folding │ ● scRNA-seq Analysis │
│ ● Sequence Alignment│ ● AlphaFold/PLM │ ● Multi-Omics Integration │
│ ● Promoter Analysis │ ● Signal Peptides │ ● Genomics + Proteomics │
│ ● Methylation (ILL) │ ● HMM Profiles │ ● Transcriptomics │
│ ● Variant Calling │ ● Domain Annotation │ ● Epigenomics │
├─────────────────────┼──────────────────────┼───────────────────────────┤
│ CLINICAL GENOMICS │ TOOLS & PIPELINES │ ML FOR BIO │
│ ● Pathogenicity Pred│ ● MEGA11 │ ● Random Forest │
│ ● ClinVar Analysis │ ● Biopython │ ● SVM / KNN │
│ ● PolyPhen / SIFT │ ● BLAST / HMMER │ ● Neural Networks │
│ ● Cancer Genomics │ ● Illumina Arrays │ ● PSO Optimization │
│ ● KCNB1 Variants │ ● Jupyter / RStudio │ ● LOOCV / Cross-Val │
└─────────────────────┴──────────────────────┴───────────────────────────┘
|
Breast Cancer Prediction
ML & DL classification of promoter DNA sequences for breast cancer prediction. Compared KNN, SVM (RBF), Neural Networks, and AdaBoost achieving 96.3% accuracy with PSO-optimized SVM.
|
ML for Genetic Variant Prediction
Random Forest model on ClinVar data to classify KCNB1 gene variants as pathogenic or benign. Benchmarked against PolyPhen and SIFT in-silico tools.
|
|
Illumina Infinium Array
Statistical analysis of fluorescent intensity data and methylation statuses from Illumina arrays using R. Covers probe characteristics, beta values, and differential methylation.
|
Protein Sequence ML Model
Predictive modeling of signal peptides in protein sequences using ML — critical for understanding protein secretion and subcellular localization.
|
|
Lab of Bioinformatics Project
Built a Profile Hidden Markov Model (pHMM) for the Kunitz-type protease inhibitor domain — a rigorous structural bioinformatics exercise using HMMER and MSA.
|
Personal Research Website
Personal portfolio built with HTML/CSS/JS showcasing research, experience, and projects in bioinformatics, AI, and computational biology.
|
| Year | Title | Journal |
|---|---|---|
| 2022 | Applications of Python Programming in Bioinformatics (Biopython) | Journal of Ghin |
| 2021 | Cancer Cell Cycle in Breast & Testicular Cancer | Journal of Ghin |
| 2020 | Targeted Drug Delivery for Cancer Treatment | Journal of Ghin |
🔬 ACTIVE RESEARCH AREAS
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🧬 Large Language Models (LLMs) for DNA/RNA sequence analysis
🔤 Protein Language Models (ESM, ProtTrans, AlphaFold integration)
🧠 Deep Reinforcement Learning for DNA sequence alignment
🕸️ Graph Neural Networks in Computational Biology
⚗️ AI + CRISPR: smart gene-editing target identification
🔬 Single-cell & Spatial Transcriptomics with Deep Learning
🌐 Multi-Omics Data Integration (Genomics + Proteomics + Transcriptomics)
🔒 Trustworthy & Interpretable AI for Clinical Genomics
🧪 Generative Models for Protein Sequence Design
🧲 Biomedical Text Mining for Drug Discovery
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
| Degree | Institution | Focus |
|---|---|---|
| MSc Bioinformatics | University of Bologna | ML, Deep Learning in Genomics, Multi-Omics, Structural Bio |
| BSc Cellular & Molecular Biology | University of Damghan | GPA: 17.87/20 · Genetics, Biostatistics, Programming |
"Decoding the language of life, one sequence at a time."