diff --git a/1_scrnaseq_data/scrnaseq_data.Rmd b/1_scrnaseq_data/scrnaseq_data.Rmd index 582a2ee55a02c6558df313b2635244a57c3585f0..b2bbf4c52220a8628cf4cfb3d93ee6293dc4c5cf 100644 --- a/1_scrnaseq_data/scrnaseq_data.Rmd +++ b/1_scrnaseq_data/scrnaseq_data.Rmd @@ -181,7 +181,7 @@ We want to estimate the number of each mRNA molecule in each cell. ## Unique Molecular Identifier (UMI) \begin{center} -\includegraphics[width=\textwidth]{./img/umi_and_pcr.png} +\includegraphics[width=\textwidth]{img/umi_and_pcr.png} \end{center} [*](https://www.genomescan.nl/wp-content/uploads/2019/10/graphic_UMIs_1.png.webp) @@ -190,7 +190,7 @@ We want to estimate the number of each mRNA molecule in each cell. ## Unique Molecular Identifier (UMI) \begin{center} -\includegraphics[width=\textwidth]{./img/umi_vs_read.png} +\includegraphics[width=\textwidth]{img/umi_vs_read.png} \end{center} ## single-cell RNA Sequencing @@ -444,7 +444,7 @@ Phred & Probability of incorrect base call & Base call accuracy \\ ## Fastq preparation \begin{center} -\includegraphics[width=0.8\textwidth]{./img/barcode_umi_sequence_split_fastq.png} +\includegraphics[width=0.8\textwidth]{img/barcode_umi_sequence_split_fastq.png} \end{center} ## Cell barcode correction @@ -457,7 +457,7 @@ Phred & Probability of incorrect base call & Base call accuracy \\ \end{block} \column{0.5\textwidth} \vspace{2em} -\includegraphics[width=0.7\textwidth]{./img/hash_function.png} +\includegraphics[width=0.7\textwidth]{img/hash_function.png} \end{columns} \end{center} @@ -466,7 +466,7 @@ Phred & Probability of incorrect base call & Base call accuracy \\ \begin{center} \begin{columns} \column{0.5\textwidth} -\includegraphics[width=0.9\textwidth]{./img/whitelist_hamming_dist.png} +\includegraphics[width=0.9\textwidth]{img/whitelist_hamming_dist.png} \column{0.5\textwidth} The list of cell barcode is known: @@ -483,7 +483,7 @@ We can take into account the sequence quality ## mRNA identification -We need to **align** the cDNA sequence part of the read to a reference genome. \includegraphics[width=0.2\textwidth]{./img/mapping_read.png} +We need to **align** the cDNA sequence part of the read to a reference genome. \includegraphics[width=0.2\textwidth]{img/mapping_read.png} \begin{block}{Smith–Waterman algorithm} Local sequence alignment, which aims at determining similar regions between two sequences. @@ -514,7 +514,7 @@ $s(a_i,b_j) = \begin{cases}+3, \quad a_i=b_j \\ -3, \quad a_i\ne b_j\end{cases}$ ## mRNA identification -We need to **align** the cDNA sequence part of the read to a reference genome. \includegraphics[width=0.2\textwidth]{./img/mapping_read.png} +We need to **align** the cDNA sequence part of the read to a reference genome. \includegraphics[width=0.2\textwidth]{img/mapping_read.png} \begin{block}{Smith–Waterman algorithm} Local sequence alignment, which aims at determining similar regions between two sequences. @@ -579,7 +579,7 @@ H_{i-1,j-1} + s(a_i,b_j), \\ \] \begin{center} -\includegraphics[width=\textwidth]{./img/Smith-Waterman-Algorithm-Example-Step1.png} +\includegraphics[width=\textwidth]{img/Smith-Waterman-Algorithm-Example-Step1.png} \end{center} ## mRNA identification @@ -589,7 +589,7 @@ H_{i-1,j-1} + s(a_i,b_j), \\ \vspace{-3.5em} \begin{center} -\includegraphics[width=0.4\textwidth]{./img/Smith-Waterman-Algorithm-Example-Step2.png} +\includegraphics[width=0.4\textwidth]{img/Smith-Waterman-Algorithm-Example-Step2.png} \end{center} We fill all the elements @@ -609,7 +609,7 @@ score of 0 \vspace{-12.5em} \begin{center} -\includegraphics[width=0.4\textwidth]{./img/Smith-Waterman-Algorithm-Example-Step3.png} +\includegraphics[width=0.4\textwidth]{img/Smith-Waterman-Algorithm-Example-Step3.png} \end{center} @@ -776,7 +776,7 @@ Do we keep reads mapping to introns ? Do we keep reads mapping after the annotated 3' UTR ? \begin{center} -\includegraphics[width=0.65\textwidth]{./img/undetected_read_intergenic_mapping.png} +\includegraphics[width=0.65\textwidth]{img/undetected_read_intergenic_mapping.png} \end{center} ## mRNA identification @@ -789,7 +789,7 @@ Do we keep reads mapping after the annotated 3' UTR ? \begin{columns} \column{0.65\textwidth} \begin{center} -\includegraphics[width=\textwidth]{./img/discared_reads.png +\includegraphics[width=\textwidth]{img/discared_reads.png } \end{center} \column{0.35\textwidth} diff --git a/2_normalization/normalization.Rmd b/2_normalization/normalization.Rmd index c750990fcdde90c559f498ee3cf59927ffb79576..3fe1a0b45c19348cfe53ee8c73f01e10a326e985 100644 --- a/2_normalization/normalization.Rmd +++ b/2_normalization/normalization.Rmd @@ -395,7 +395,7 @@ With a transcription rate $\lambda_g(t)$ the observed mRNA count follow a Poisso $P(X = x)$ for $\mathcal{P}(\lambda_g)$ \begin{center} -\includegraphics[width=0.6\textwidth]{./img/poisson.png} +\includegraphics[width=0.6\textwidth]{img/poisson.png} \end{center} The expectation of $X$, $E(X)$ is equal to it's variance $Var(X)$ (both are equal to $\lambda_g$) @@ -467,7 +467,7 @@ In bulk RNASeq, we have $\sim 3$ observation per gene the task is more difficult $P(X = x)$ for $\mathcal{P}(\mu)$ \begin{center} -\includegraphics[width=0.6\textwidth]{./img/poisson.png} +\includegraphics[width=0.6\textwidth]{img/poisson.png} \end{center} \begin{center} @@ -513,7 +513,7 @@ with $Var(X) = \lambda + \alpha \lambda^2$ \column{0.5\textwidth} \vspace{1em} -\includegraphics[width=0.9\textwidth]{./img/mu_vs_var.png} +\includegraphics[width=0.9\textwidth]{img/mu_vs_var.png} \end{columns} \end{center} @@ -523,7 +523,7 @@ with $Var(X) = \lambda + \alpha \lambda^2$ $P(X = x)$ for $\mathcal{P}(\lambda)$ \begin{center} -\includegraphics[width=0.8\textwidth]{./img/poisson.png} +\includegraphics[width=0.8\textwidth]{img/poisson.png} \end{center} @@ -532,7 +532,7 @@ $P(X = x)$ for $\mathcal{P}(\lambda)$ $P(X = x)$ for $\mathcal{NB}(\lambda, \alpha = 10)$ \begin{center} -\includegraphics[width=0.8\textwidth]{./img/NB_sigma_10.png} +\includegraphics[width=0.8\textwidth]{img/NB_sigma_10.png} \end{center} @@ -541,7 +541,7 @@ $P(X = x)$ for $\mathcal{NB}(\lambda, \alpha = 10)$ $P(X = x)$ for $\mathcal{NB}(\lambda, \alpha = 2)$ \begin{center} -\includegraphics[width=0.8\textwidth]{./img/NB_sigma_2.png} +\includegraphics[width=0.8\textwidth]{img/NB_sigma_2.png} \end{center} ## Counts distributions @@ -549,7 +549,7 @@ $P(X = x)$ for $\mathcal{NB}(\lambda, \alpha = 2)$ $P(X = x)$ for $\mathcal{NB}(\lambda, \alpha = 1)$ \begin{center} -\includegraphics[width=0.8\textwidth]{./img/NB_sigma_1.png} +\includegraphics[width=0.8\textwidth]{img/NB_sigma_1.png} \end{center} ## Variance of count data @@ -867,7 +867,7 @@ Sanity will estimate $log\left(\alpha_{gi}\right) \sim \mathcal{N}\left(\mu, \si ## Goals of Normalization \begin{center} -\includegraphics[width=\textwidth]{./img/normalization_marker_gene.png} +\includegraphics[width=\textwidth]{img/normalization_marker_gene.png} \end{center} @@ -876,7 +876,7 @@ Sanity will estimate $log\left(\alpha_{gi}\right) \sim \mathcal{N}\left(\mu, \si ## batch effects \begin{center} -\includegraphics[width=\textwidth]{./img/batch_effect.png} +\includegraphics[width=\textwidth]{img/batch_effect.png} \end{center} **batch** effects appear when you mix data from different: @@ -906,7 +906,7 @@ We can run Sanity on each separate batch and combine the results. ## Harmony \begin{center} -\includegraphics[width=\textwidth]{./img/harmony.png} +\includegraphics[width=\textwidth]{img/harmony.png} \end{center} # single-cell RNA-Seq dimension reduction *Monday 13 June 2022* diff --git a/3_dimension_reduction/dimension_reduction.Rmd b/3_dimension_reduction/dimension_reduction.Rmd index 42cd5085b2de32da3736c7facb2db1363d245f53..47137cb16656097fcc9213745fd46149f8ed35b9 100644 --- a/3_dimension_reduction/dimension_reduction.Rmd +++ b/3_dimension_reduction/dimension_reduction.Rmd @@ -275,7 +275,7 @@ Dimension reduction is mandatory for any analysis (clustering, visualization, in \vspace{-2em} \begin{center} - \includegraphics[height=4cm]{./img/matrix_factorization.png} + \includegraphics[height=4cm]{img/matrix_factorization.png} \end{center} \begin{center} @@ -495,7 +495,7 @@ x_{2,i} \\ \vspace{1em} \begin{center} - \includegraphics[height=4cm]{./img/sparce_matrix_factorization.png} + \includegraphics[height=4cm]{img/sparce_matrix_factorization.png} \end{center} @@ -504,7 +504,7 @@ x_{2,i} \\ ### Non-negative matrix factorization \begin{center} \href{https://doi.org/10.1093/bioinformatics/btz177}{ - \includegraphics[width=\textwidth]{./img/count_matrix_factorization.png} + \includegraphics[width=\textwidth]{img/count_matrix_factorization.png} } \end{center} \vspace{-2em} diff --git a/5_pseudo_time/pseudo_time.Rmd b/5_pseudo_time/pseudo_time.Rmd index 186f5dc70a517b791bdaeeeecd403a45a26c333e..178b511469a071cc2f452bdc4bd2f1198ecfd761 100644 --- a/5_pseudo_time/pseudo_time.Rmd +++ b/5_pseudo_time/pseudo_time.Rmd @@ -137,6 +137,7 @@ one RNA-Seq experiment constitutes a {\bf time series}, with each {\bf cell repr ## Monocle 1 + \begin{center} \href{http://www.nature.com/doifinder/10.1038/nbt.2859}{ \includegraphics[width=\textwidth]{img/monocle_1.png} @@ -146,6 +147,10 @@ one RNA-Seq experiment constitutes a {\bf time series}, with each {\bf cell repr ## Monocle 1 +### Find the longest path + +\vspace{-2em} + \begin{center} \href{http://www.nature.com/doifinder/10.1038/nbt.2859}{ \includegraphics[width=0.6\textwidth]{img/monocle_2.png} @@ -210,10 +215,10 @@ traveling salesman problem (TSP) \end{center} \column{0.4\textwidth} \begin{itemize} - \item a. The linear regression line - \item b. The principal-component - \item c. The smooth regression curve - \item d. The principal curve minimizes + \item linear regression line + \item principal-component + \item smooth regression curve + \item principal curve minimizes \end{itemize} \end{columns} \end{center}