Here reveals the mathematical foundations of **latent time**, which enables in-depth grasp and interpretation of the latent time of RNA velocity. This is the fourth installment of our blog series on effectively applying the dynamic model to infer RNA velocity from single-cell RNA-seq.
In this third blog on effectively applying the dynamic model of RNA velocity, we look into post hoc computed cosine similarity and the exponential kernel that shape the RNA velocity graph and embedding. This begins our deep dive into scVelo’s post hoc computations that determine visualization and interpretation.
Here derives the mathematics underpinning the parameter inference of dynamic RNA velocity model, which is the second installment of our blog series to effectively apply the dynamic model in revealing the RNA velocity of single-cell RNAseq.
Here we delve into the mathematical foundations of the dynamic model of RNA velocity. This is the 1st installment of our blog series to clearly understand strengths and limitations of the dynamic model that is required for its effective application.
Mathematical models used in SAVER, a tool of scRNAseq imputation, could well depict the structure inherent to the scRNAseq datasets as suggested by its superior performance. Here we derive SAVER's Poisson–gamma mixture model (also known as negative binomial model) and its Bayesian framework that leverage conjugate priors to estimate the posterior distribution of gene expression levels.
Human Endogenous Retroviruses (hERVs) are ancient viral sequences embedded in the human genome. Transgenes are common in transgenic mouse models. To quantify them from sequencing reads, we need-- a) modify fasta and gtf files to include the their sequencing and annotation; b) a feature quantification algorithm to handle multimapping commonly seen for hERV and transgenes. Here I discussed the algorithms for feature quantification, and successfully quantified hERV and transgenes by implementing an EM algorithm.
sctransfrom is highly recommened in spatial genomic analysis, and was first introduced to analyze scRNAseq. Here, we deciper it and reveal what it is really doing mathmatically to have a sense of its applicability.
Well based scRNAseq has many more details than 10x scRNAseq to pay attention to in order to quantify features using Starsolo. Here, we discuss and present the snakemake pipline to quantify hERV, transgenes and genomics from well-based single-cell RNAseq data
A snakemake pipline to quantify hERV, transgenes and genomics from 10x single-cell RNAseq data
ERV gtf annotation is needed to quantify ERV from sequencing data. But the ERV gtf compatible with the latest mouse genome GRCm39 is not publicly available. Here I generated and validated the ERV gtf based on GRCm39 using RepeatMasker and customized scripts.
