Projects
Deep Learning
- Single-cell-RNA-seq and Single-nuclei-RNA-seq data Integration using neural networks structures for dimensionality reduction and further downstream analysis processing.
- Data representation and embedding tuning using UMAP, PYMDE, PCA, K-Means…
- Face emotion recognition mini-project “for fun” to test and set up my local resources (GPU, SDD, and RAM).
The projects were executed using different libraries in Python.
Comparisons of RNA-seq data from Danio rerio heart injury models
The workflow established to analyze the RNA-seq gathered data can be understood as a data processing pipeline that cleans variables not related to the biological question of interest and adds information to the genes (data of interest) obtained after the analysis to end up with a Biological Interpretation.
Biological Interpretation is helped by multiple visualizations such as Integrated Networks realized with Gene Ontologies obtained through the use of diverse tools.
The R code files that were used for this analysis will be found on GitHub.
small RNA-seq species data from Danio rerio analysis
Capture the micro, piwi, tRNA, rRNA, and more to add species to the workflow established using remote resources via SLURM.
Variant Calling for RNA-seq data from Danio rerio for capturing the Haplotypes involved
Metabolic disturbances it is believed could arise from variants affecting the mitochondrion in charge of the energy supply of the cell and altering completely the organism.
Genomic data processing
Using remote resources such as “Clusters” for processing data with different tools such as FASTQC, MultiQ, Cutadap, STAR, CellRanger, BWA, and Samtools. For further processing: Seurat(R), Scanpy(Python), scVI, clusterPorfiler, GSEA, GO, and many more to analyze the data from raw to end.
Shiny web apps
Development of web applications, designed for interactive data visualization and downloading. The web applications are customizable, versatile, secure, and based on the R programming language. Example (requires password and user)