I’m answering this assuming you mean bulk RNA-seq. If you mean scRNA-seq instead, several of these resources are still relevant, but you’ll want to use a search engine or LLM for scRNA-seq-specific additions.

Although it’s not a course per se, I think the regularly updated DESeq2 tutorial is a solid foundation for much of what you’re asking (experimental design, design formulas, modeling, and practical QC/EDA), especially for design formulas and thinking around GLMs (interactions, likelihood ratio tests [LRTs], contrasts, etc.), and the modeling workflow in general. What’s nice is that it also covers a lot of the “why” and the day-in-day-out mechanics leading up to the models (normalization/size factors, mean-variance behavior and dispersion, shrinkage, and LFC interpretation), plus standard diagnostic plots (MA, PCA, sample distances, dispersion trends). Still, it doesn’t go deep on all fundamentals.

In a similar vein, there are some excellent publicly available RNA-seq differential expression lecture materials/notes/slides floating around (e.g., course-style, stats-focused as opposed to scripting-focused). For example, if you search for “Introduction to differential gene expression analysis using RNA-seq” you’ll find a widely shared, very detailed PDF on this material (e.g., here). If you find that PDF useful, you can plug the authors’ names with “rna-seq” into a search engine to turn up more resources, including course syllabuses and lecture materials.

For more “course-like” RNA-seq narratives, the Bioconductor rnaseqGene workflow and related Bioconductor/CSAMA course materials are great “counts to design to modeling to QC/EDA” walkthroughs, and they tend to keep the practical work grounded in the statistical framing. Harvard Chan’s HBC bulk RNA-seq workshop workshop is also course-like, approachable, and covers many of the topics you listed. For the above, some lessons do involve scripting, but the material is modular enough that it’s easy to skim/skip the tooling parts (since it appears to not be your focus).

For “complex designs” and “time course analysis:” DESeq2 can handle many time-course questions via interactions (time × condition) and/or LRTs (e.g., see here), and so can limma-voom and edgeR, so it’s worth studying those resources too. In practice, time course analysis is not really about “one true method;” instead, it’s about being explicit about the question (e.g., difference at specific time points, difference in trends, overall condition effect while accounting for time, etc.) and then choosing an appropriate model and contrasts. In particular, the edgeR user guide (and, more broadly, material from those authors), is very detailed and granular on experimental design and linear modeling.

Regarding “coverage needed,” this tends to be a complicated question and is highly dependent on the study question and design. A rough rule of thumb is that biological replication and controlling for unwanted variation often lead to increased statistical power versus increasing sequencing depth beyond a reasonable baseline (and even defining “reasonable baseline” depends on a lot of factors). The right tradeoff depends on effect sizes, variance, tissue complexity, and whether you care about lowly expressed genes and/or isoforms (among other things). If you can, a small pilot (or using public data for [or close to] your system) to estimate variability is often the most practical way to reason about power. Power tools can help too, but they’re only as good as the assumptions they’re fed. For more details, you can use a search engine for, e.g., “statistical power for rna-seq” and peruse what comes up (some of it is educational in nature, even if it’s not exactly packaged as a course).

For math and modeling that is not RNA-seq-specific, “Introduction to Statistical Learning” (R or Python) is a great general on-ramp (e.g., especially for GLMs). If you want something that’s more explicitly aligned with biological data analysis, “Modern Statistics for Modern Biology” is excellent and (in my experience) matches how many analysts think about models/experiments day-in and day-out without drowning you in first-principles derivations (that said, it doesn’t hit the fundamentals like dedicated statistics textbooks do [which is something you may want]). Both are freely online and have associated lecture material.

Finally, the “Biostars Handbook” is a good practical reference alongside the more “model-first” resources; it covers many of the topics you listed (and a lot more beyond that). Same with ngs101.com (a newer site), and you can’t go wrong with materials from Ming Tommy Tang (e.g., as in here). I also like “Statistics by Jim” (Jim Frost) for building intuition around regression and GLMs, even if it’s not a core “RNA-seq design” resource; for example, here’s a nice post on choosing an appropriate regression. I recommend his books too; they’re not expensive, and they may be discounted for people in low-income countries. (I think this is the same for Ming Tommy Tang’s book, which I also recommend, although it covers much more than the topics you listed.)

Looking over this answer, there are many keywords you can pull out and use in search engines and LLMs for even more material. Wishing you the best on this (long) journey.

We have a course designed for beginners: RNASEQ DATA ANALYSIS FOR BEGINNERS

And one focused on the statistical analyses: RNA SEQUENCING DATA WITH R/BIOCONDUCTOR

Take a look at Data Analysis for Life Sciences (includes a course on Linear Models & Matrix Algebra, and on Statistical Inference & Modeling for High-Throughput Experiments )

and

Statistics for Genomic Data Science