China 
Africa 
Explore chapters and articles related to this topic
Big Data and Transcriptomics
Published in Shampa Sen, Leonid Datta, Sayak Mitra, Machine Learning and IoT, 2018
Sudharsana Sundarrajan, Sajitha Lulu, Mohanapriya Arumugam
The high-throughput technologies provide a valuable pathway to study the transcripts of organisms. The technologies provide more insight about the genes expressed in different disease condition, disease progression, RNA editing, fusion transcripts, alternative splicing, and allele-specific expression. Transcriptomics research is moving forward towards single-cell transcriptomics studies, since the expression of the genes vary from one cell to other. Mining data from public repositories such as ENCODE, TCGA, GEO, and the Geuvadis project may provide new insight about gene regulation. The high volume of data requires high-end computational resources for transcriptome assembly and other downstream analysis. Hence, highly parallel data processing steps are required for better analysis and faster processing of the results.
Joint L2,p-norm and random walk graph constrained PCA for single-cell RNA-seq data
Published in Computer Methods in Biomechanics and Biomedical Engineering, 2023
Tai-Ge Wang, Jun-Liang Shang, Jin-Xing Liu, Feng Li, Shasha Yuan, Juan Wang
Identifying cell types is an unsupervised clustering problem. Improving the performance of single-cell clustering is vital for accurately identifying cell types. However, single-cell transcriptomics suffers from high levels of technical noise, resulting in only a small fraction of the transcriptome of each cell being captured during amplification. Moreover, the high dimensionality of scRNA-seq data makes it more difficult to apply traditional clustering methods directly (Von Luxburg 2007; Murtagh and Contreras 2012). Validation of cell type identification accuracy based on high-dimensional, high-noise scRNA-seq data remains a challenge (Peng et al. 2019).