Biology subjects

Cabrero, D. G.

Publications and source records attributed to Cabrero, D. G..

3 recordsLinked to original sources

bioRxiv2023

HLA-Based Banking of Human Induced Pluripotent Stem Cells in Saudi Arabia

Human iPSCs derivation and use in clinical studies are transforming medicine. Yet, there is a high cost and long waiting time for autologous iPS-based cellular therapy, and the genetic engineering of hypo-immunogenic iPS cell lines is hampered with numerous hurdles. Therefore, it is increasingly interesting to create cell stocks based on HLA haplotype distribution in a given population. In this study, we assessed the potential of HLA-based iPS banking for the Saudi population. First, we analyzed the HLA database of the Saudi Stem Cell Donor Registry (SSCDR), which contains high-resolution HLA genotype data of 64,315 registered Saudi donors at the time of analysis. We found that only 13 iPS lines would be required to cover 30% of the Saudi population, 39 iPS lines would offer 50% coverage and 596 for more than 90% coverage. Next, As a proof-of-concept, we launched the first HLA-based banking of iPSCs in Saudi Arabia. Using clinically relevant methods, we generated the first iPSC line from a homozygous donor for the most common HLA haplotype in Saudi. The two generated clones expressed pluripotency markers, could be differentiated into all three germ layers, beating cardiomyocytes and neuronal progenitors. To ensure that our reprogramming method generates genetically stable iPSCs, we assessed the mutational burden in the generated clones and the original blood sample from which the iPSCs were derived using whole-genome sequencing. All detected variants were found in the original donor sample and were classified as benign according to current guidelines of the American College of Medical Genetics and Genomics (ACMG). This study sets a road map for introducing iPS-based cell therapy in the Kingdom of Saudi Arabia.

genetics↗

bioRxiv2023

LEP-AD: Language Embedding of Proteins and Attention to Drugs predicts drug target interactions

AO_SCPLOWBSTRACTC_SCPLOWPredicting drug-target interactions is a tremendous challenge for drug development and lead optimization. Recent advances include training algorithms to learn drug-target interactions from data and molecular simulations. Here we utilize Evolutionary Scale Modeling (ESM-2) models to establish a Transformer protein language model for drug-target interaction predictions. Our architecture, LEP-AD, combines pre-trained ESM-2 and Transformer-GCN models predicting binding affinity values. We report new best-in-class state-of-the-art results compared to competing methods such as SimBoost, DeepCPI, Attention-DTA, GraphDTA, and more using multiple datasets, including Davis, KIBA, DTC, Metz, ToxCast, and STITCH. Finally, we find that a pre-trained model with embedding of proteins (the LED-AD) outperforms a model using an explicit alpha-fold 3D representation of proteins (e.g., LEP-AD supervised by Alphafold). The LEP-AD model scales favorably in performance with the size of training data. Code available at https://github.com/adaga06/LEP-AD

bioinformatics↗

bioRxiv2022

scAEGAN: Unification of Single-Cell Genomics Data by Adversarial Learning of Latent Space Correspondences

Recent progress in Single-Cell Genomics have produced different library protocols and techniques for profiling of one or more data modalities in individual cells. Machine learning methods have separately addressed specific integration challenges (libraries, samples, paired-unpaired data modalities). We formulate an unifying data-driven methodology addressing all these challenges. To this end, we design a hybrid architecture using an autoencoder (AE) network together with adversarial learning by a cycleGAN (cGAN) network, jointly referred to as scAEGAN. The AE learns a low-dimensional embedding of each condition, whereas the cGAN learns a non-linear mapping between the AE representations. The core insight is that the AE respects each samples uniqueness, whereas the cGAN exploits the distributional data similarity in the latent space. We evaluate scAEGAN using simulated data and real datasets of a single-modality (scRNA-seq), different library preparations (Fluidigm C1, CelSeq, CelSeq2, SmartSeq), and several data modalities such as paired scRNA-seq and scATAC-seq. We find that scAEGAN outperforms Seurat3 in library integration, is more robust against data sparsity, and beats Seurat 4 in integrating paired data from the same cell. Furthermore, in predicting one data modality from another, scAEGAN outperforms Babel. We conclude scAEGAN surpasses current state-of-the-art methods across several seemingly different integration challenges.

bioinformatics↗