Elena Dalboni da Rocha · @enterlina
Design & Product Portfolio · Machine Learning · AI Engineering
A computational genomics portfolio documenting 9 Python and Jupyter implementations from my bioinformatics studies. This work focuses on analyzing DNA sequencing reads, understanding errors introduced during sequencing and alignment, evaluating correction methods, and constructing graphs for genome assembly.
Core areas: FASTQ quality control · DNA read alignment · genomic coverage · insertion/deletion analysis · long-read sequencing · error correction · de Bruijn graphs
FASTQ reads → base quality & GC analysis → SAM/BAM alignment → coverage & error profiles → correction assessment → de Bruijn graph assembly experiments
| Topic | What the code explores | Source |
|---|---|---|
| 01 · FASTQ quality | FASTQ read QC: GC-content distributions under quality filters and per-base Phred quality boxplots. | 01-fastq-gc-content-and-phred-quality.py |
| 01 · FASTQ quality | Biopython FASTQ parsing, quality-aware GC histograms, per-position quality and FastQC comparison. | 01-fastq-quality-control-notebook.ipynb |
| 02 · Alignment analysis | SAM/BAM alignment depth via difference-array prefix sums; insert-size distribution and reference substitution counts. | 02-alignment-coverage-insert-size-substitutions.py |
| 02 · Error profiling | Aligned-read substitutions and indels, coverage, error percentages, homopolymer length discrepancies and base-quality comparisons. | 02-sequencing-error-profiles-indels-homopolymers.py |
| 03 · Long-read sequencing | Comparison of PacBio and Oxford Nanopore alignments using genome coverage, substitution tables and indel length profiles. | 03-long-read-platform-coverage-error-comparison.py |
| 04 · Error correction | Base-level comparison of raw versus corrected alignments, partitioning true corrections, missed errors, false corrections and removed bases. | 04-read-error-correction-evaluation.py |
| 05 · Genome assembly | k-mer / reverse-complement De Bruijn graph construction, coverage-aware edge compaction, tip simplification and graph exports. | 05-debruijn-graph-genome-assembly-notebook.ipynb |
| 05 · Genome assembly | Graph vertex/edge construction from sequence reads, reverse complements, compaction and DOT/PNG visualization. | 05-debruijn-graph-construction-visualization.py |
| 05 · Genome assembly | Extended genome assembly graph operations, path compaction, low-coverage/short-tip removal and sequence export. | 05-debruijn-graph-tip-removal-assembly.py |
- Sequence data processing: Parsing FASTQ/FASTA formats, Phred scores, nucleotides and reverse complements with Biopython and pysam.
- Alignment analytics: Per-position coverage using a difference array and prefix sums; insert-size histograms; substitution and indel matrices.
- Sequencing technology comparison: Code for evaluating read-alignment error distributions from PacBio and Oxford Nanopore datasets.
- Error-correction audit: Classifying bases as corrected, uncorrected, falsely modified or removed by comparing aligned reads.
- Graph algorithms for genome assembly: Building a k-mer De Bruijn graph, maintaining directed edges and read coverage, merging non-branching paths, pruning tips and exporting graph visualizations.
Python · Biopython · pysam · NumPy · pandas · Matplotlib · seaborn · Jupyter · PyGraphviz
These are historical algorithm exercises, not a clinical diagnostic pipeline or production-validated genome assembler. The original FASTQ/FASTA/SAM/BAM datasets and evaluation outputs are not included in the repository. Several scripts use absolute local paths and legacy dependencies; not every implementation has been tested for current Python versions. Code is preserved as authored; the descriptive filenames and this index do not imply that all analyses run end-to-end or have validated accuracy.
The repository also retains historical IDE and virtual-environment artifacts for provenance; they are not recommended as a reproducible environment.
Elena Dalboni da Rocha — AI Product Designer, Design Engineer and bioinformatics graduate, combining computational biology foundations with modern AI product development.