Add library-level and sample-level RNA metrics to report

public/gentrap/src/main/resources/nl/lumc/sasc/biopet/pipelines/gentrap/scripts/pdf_report.py

@@ -419,6 +419,31 @@ class GentrapLib(object):
             self.fastqc_r2_qc = FastQC(self.fastqc_r2_qc_files["fastqc_data"]["path"])
         # mapping metrics settings
         self.aln_metrics = summary.get("bammetrics", {}).get("stats", {}).get("alignment_metrics", {})
+        # insert size metrics files
+        self.inserts_metrics_files = summary.get("bammetrics", {}).get("files", {}).get("insert_size_metrics", {})
+        # rna metrics files and stats
+        _rmetrics = summary.get("gentrap", {}).get("stats", {}).get("rna_metrics", {})
+        if _rmetrics:
+            self.rna_metrics_files = summary.get("gentrap", {}).get("files", {}).get("rna_metrics", {})
+            self.rna_metrics = {k: v for k, v in _rmetrics.items() }
+            pf_bases = float(_rmetrics["pf_bases"])
+            exonic_bases = int(_rmetrics.get("coding_bases", 0)) + int(_rmetrics.get("utr_bases", 0))
+            # picard uses pct_ but it's actually ratio ~ we follow their convention
+            pct_exonic_bases_all = exonic_bases / float(_rmetrics["pf_bases"])
+            pct_exonic_bases = exonic_bases / float(_rmetrics.get("pf_aligned_bases", 0))
+            self.rna_metrics.update({
+                    "exonic_bases": exonic_bases,
+                    "pct_exonic_bases_all": pct_exonic_bases_all,
+                    "pct_exonic_bases": pct_exonic_bases,
+                    "pct_aligned_bases": 1.0,
+                    "pct_aligned_bases_all": float(_rmetrics.get("pf_aligned_bases", 0.0)) / pf_bases,
+                    "pct_coding_bases_all": float(_rmetrics.get("coding_bases", 0.0)) / pf_bases,
+                    "pct_utr_bases_all": float(_rmetrics.get("utr_bases", 0.0)) / pf_bases,
+                    "pct_intronic_bases_all": float(_rmetrics.get("intronic_bases", 0.0)) / pf_bases,
+                    "pct_intergenic_bases_all": float(_rmetrics.get("intergenic_bases", 0.0)) / pf_bases,
+            })
+            if _rmetrics.get("ribosomal_bases", "") != "":
+                self.rna_metrics["pct_ribosomal_bases_all"] = float(_rmetrics.get("pf_ribosomal_bases", 0.0)) / pf_bases
 
     def __repr__(self):
         return "{0}(sample=\"{1}\", lib=\"{2}\")".format(
@@ -436,6 +461,30 @@ class GentrapSample(object):
         self.libs = \
             {l: GentrapLib(self.run, self, l, summary["libraries"][l]) \
                 for l in self.lib_names}
+        # rna metrics files and stats
+        _rmetrics = summary.get("gentrap", {}).get("stats", {}).get("rna_metrics", {})
+        if _rmetrics:
+            self.rna_metrics_files = summary.get("gentrap", {}).get("files", {}).get("rna_metrics", {})
+            self.rna_metrics = {k: v for k, v in _rmetrics.items() }
+            pf_bases = float(_rmetrics["pf_bases"])
+            exonic_bases = int(_rmetrics.get("coding_bases", 0)) + int(_rmetrics.get("utr_bases", 0))
+            # picard uses pct_ but it's actually ratio ~ we follow their convention
+            pct_exonic_bases_all = exonic_bases / float(_rmetrics["pf_bases"])
+            pct_exonic_bases = exonic_bases / float(_rmetrics.get("pf_aligned_bases", 0))
+            self.rna_metrics.update({
+                    "exonic_bases": exonic_bases,
+                    "pct_exonic_bases_all": pct_exonic_bases_all,
+                    "pct_exonic_bases": pct_exonic_bases,
+                    "pct_aligned_bases": 1.0,
+                    "pct_aligned_bases_all": float(_rmetrics.get("pf_aligned_bases", 0.0)) / pf_bases,
+                    "pct_coding_bases_all": float(_rmetrics.get("coding_bases", 0.0)) / pf_bases,
+                    "pct_utr_bases_all": float(_rmetrics.get("utr_bases", 0.0)) / pf_bases,
+                    "pct_intronic_bases_all": float(_rmetrics.get("intronic_bases", 0.0)) / pf_bases,
+                    "pct_intergenic_bases_all": float(_rmetrics.get("intergenic_bases", 0.0)) / pf_bases,
+            })
+            if _rmetrics.get("ribosomal_bases", "") != "":
+                self.rna_metrics["pct_ribosomal_bases_all"] = float(_rmetrics.get("pf_ribosomal_bases", 0.0)) / pf_bases
+
 
     def __repr__(self):
         return "{0}(\"{1}\")".format(self.__class__.__name__, self.name)

public/gentrap/src/main/resources/nl/lumc/sasc/biopet/pipelines/gentrap/templates/pdf/lib_mapping.tex

 \subsection{Mapping}
-\label{sec:map}
+\label{sec:map-((( lib.sample.name )))-((( lib.name )))}
 
 \subsubsection{Mapping statistics}
 
-Table~\ref{tab:bamstat-((( lib.sample.name )))-((( lib.name ))))} shows the overview statistics of the
-generated alignment file.
-
 \indent
 
 % number + percentage of reads mapped to genome
 % number + percentage of properly paired reads
 \begin{center}
     \captionof{table}{Mapping Overview}
-    \label{tab:bamstat-((( lib.sample.name )))-((( lib.name ))))}
+    \label{tab:bamstat-((( lib.sample.name )))-((( lib.name )))}
     \setlength{\tabcolsep}{11pt}
     ((* if lib.is_paired_end *))
     \begin{tabular}{ l r r r }
@@ -46,34 +43,19 @@ generated alignment file.
     \end{tabular}
 \end{center}
 
-((=
-%((* if run.is_paired_end *))
-%% inferred insert size distribution
-%\subsection{Insert size distribution}
-%
-%This section contains a quick overview of the insert size distribution
-%of the mapped reads.
-%
-%\indent
-%
-%There are three possible types of inserts the graph may denote:
-%\begin{description}
-%    \item[\textit{inward}] \hfill \\
-%        Each read pair maps to opposite strands and point towards each other.
-%    \item[\textit{outward}] \hfill \\
-%        Each read pair maps to opposite strands and point away from each other.
-%    \item[\textit{same directions}] \hfill \\
-%        Both read pair map to the same strand.
-%\end{description}
-%
-%\IfFileExists{((( vars['OUT_DIR'] )))/((( vars['SAMPLE'] ))).insertsizes.png}
-%{
-%    \begin{figure}[h!]
-%        \centering
-%        \includegraphics[width=0.7\textwidth]{((( vars['OUT_DIR'] )))/((( vars['SAMPLE'] ))).insertsizes.png}
-%        \caption{Distribution of insert size length of paired-end reads mapped to opposite strands.}
-%    \end{figure}
-%}
+((* if lib.is_paired_end *))
+% inferred insert size distribution
+\subsubsection{Insert size distribution}
+
+\IfFileExists{((( lib.inserts_metrics_files.output_histogram.path )))}
+{
+    \begin{figure}[h!]
+        \centering
+        \includegraphics[width=0.7\textwidth]{((( lib.inserts_metrics_files.output_histogram.path )))}
+        \caption{Distribution of insert size length of paired-end reads mapped to opposite strands.}
+    \end{figure}
+}
+((= TODO: strand-specific stats
 %{
 %    \IfFileExists{((( vars['OUT_DIR'] )))/((( vars['SAMPLE'] ))).f.insertsizes.png}
 %    {
@@ -92,5 +74,42 @@ generated alignment file.
 %        \end{figure}
 %    }{}
 %}
-%((* endif *))
 =))
+((* endif *))
+
+((* if lib.sample.libs|length > 1 *))
+\subsubsection{RNA-specific metrics}
+
+\IfFileExists{((( lib.rna_metrics_files.output_chart.path )))}
+{
+    \begin{figure}[h!]
+        \centering
+        \includegraphics[width=0.7\textwidth]{((( lib.rna_metrics_files.output_chart.path )))}
+        \caption{Normalized coverage bias plot.}
+    \end{figure}
+}
+
+\begin{center}
+    \captionof{table}{Functional annotation metrics}
+    \label{tab:fannot-((( lib.sample.name )))-((( lib.name ))))}
+    \setlength{\tabcolsep}{11pt}
+    \begin{tabular}{ l r r r }
+        \hline
+        \multirow{2}{*}{Parameter} & \multicolumn{3}{c}{Value} \\
+                                   & Count & \% of all & \% of aligned \\
+        \hline \hline
+        Total bases & ((( lib.rna_metrics.pf_bases|nice_int ))) & 100\% & - \\
+        Aligned bases & ((( lib.rna_metrics.pf_aligned_bases|nice_int ))) & ((( lib.rna_metrics.pct_aligned_bases_all|float2nice_pct )))\% & ((( lib.rna_metrics.pct_aligned_bases|float2nice_pct )))\% \\
+        Exonic bases & ((( lib.rna_metrics.exonic_bases|nice_int ))) & ((( lib.rna_metrics.pct_exonic_bases_all|float2nice_pct )))\% & ((( lib.rna_metrics.pct_exonic_bases|float2nice_pct )))\% \\
+            \hspace*{4mm}Coding bases & ((( lib.rna_metrics.coding_bases|nice_int ))) & ((( lib.rna_metrics.pct_coding_bases_all|float2nice_pct )))\% & ((( lib.rna_metrics.pct_coding_bases|float2nice_pct )))\% \\
+            \hspace*{4mm}UTR bases & ((( lib.rna_metrics.utr_bases|nice_int ))) & ((( lib.rna_metrics.pct_utr_bases_all|float2nice_pct )))\% & ((( lib.rna_metrics.pct_utr_bases|float2nice_pct )))\% \\
+        Intronic bases & ((( lib.rna_metrics.intronic_bases|nice_int ))) & ((( lib.rna_metrics.pct_intronic_bases_all|float2nice_pct )))\% & ((( lib.rna_metrics.pct_intronic_bases|float2nice_pct )))\% \\
+        Intergenic bases & ((( lib.rna_metrics.intergenic_bases|nice_int ))) & ((( lib.rna_metrics.pct_intergenic_bases_all|float2nice_pct )))\% & ((( lib.rna_metrics.pct_intergenic_bases|float2nice_pct )))\% \\
+        ((* if lib.rna_metrics.ribosomal_bases != "" *))
+        Ribosomal bases & ((( lib.rna_metrics.ribosomal_bases|nice_int ))) & ((( lib.rna_metrics.pct_ribosomal_bases_all|float2nice_pct )))\% & ((( lib.rna_metrics.pct_ribosomal_bases|float2nice_pct )))\% \\
+        ((* endif *))
+        \hline
+    \end{tabular}
+\end{center}
+
+((* endif *))

public/gentrap/src/main/resources/nl/lumc/sasc/biopet/pipelines/gentrap/templates/pdf/sample.tex

 \part{Sample "((( sample.name )))" Results}
 \label{sample:(((sample.name)))}
 
-Hello from module ((( sample ))) \\
+((* include "sample_mapping.tex" *))
 
 ((* for lib in sample.libs.values() *))
 ((* include "lib.tex" *))

public/gentrap/src/main/resources/nl/lumc/sasc/biopet/pipelines/gentrap/templates/pdf/sample_mapping.tex

+\section{Mapping}
+\label{sec:map-((( sample.name )))}
+
+\subsection{Mapping statistics}
+
+\indent
+
+
+\subsection{RNA-specific metrics}
+
+\IfFileExists{((( sample.rna_metrics_files.output_chart.path )))}
+{
+    \begin{figure}[h!]
+        \centering
+        \includegraphics[width=0.7\textwidth]{((( sample.rna_metrics_files.output_chart.path )))}
+        \caption{Normalized coverage bias plot.}
+    \end{figure}
+}
+
+\begin{center}
+    \captionof{table}{Functional annotation metrics}
+    \label{tab:fannot-((( sample.name )))}
+    \setlength{\tabcolsep}{11pt}
+    \begin{tabular}{ l r r r }
+        \hline
+        \multirow{2}{*}{Parameter} & \multicolumn{3}{c}{Value} \\
+                                   & Count & \% of all & \% of aligned \\
+        \hline \hline
+        Total bases & ((( sample.rna_metrics.pf_bases|nice_int ))) & 100\% & - \\
+        Aligned bases & ((( sample.rna_metrics.pf_aligned_bases|nice_int ))) & ((( sample.rna_metrics.pct_aligned_bases_all|float2nice_pct )))\% & ((( sample.rna_metrics.pct_aligned_bases|float2nice_pct )))\% \\
+        Exonic bases & ((( sample.rna_metrics.exonic_bases|nice_int ))) & ((( sample.rna_metrics.pct_exonic_bases_all|float2nice_pct )))\% & ((( sample.rna_metrics.pct_exonic_bases|float2nice_pct )))\% \\
+            \hspace*{4mm}Coding bases & ((( sample.rna_metrics.coding_bases|nice_int ))) & ((( sample.rna_metrics.pct_coding_bases_all|float2nice_pct )))\% & ((( sample.rna_metrics.pct_coding_bases|float2nice_pct )))\% \\
+            \hspace*{4mm}UTR bases & ((( sample.rna_metrics.utr_bases|nice_int ))) & ((( sample.rna_metrics.pct_utr_bases_all|float2nice_pct )))\% & ((( sample.rna_metrics.pct_utr_bases|float2nice_pct )))\% \\
+        Intronic bases & ((( sample.rna_metrics.intronic_bases|nice_int ))) & ((( sample.rna_metrics.pct_intronic_bases_all|float2nice_pct )))\% & ((( sample.rna_metrics.pct_intronic_bases|float2nice_pct )))\% \\
+        Intergenic bases & ((( sample.rna_metrics.intergenic_bases|nice_int ))) & ((( sample.rna_metrics.pct_intergenic_bases_all|float2nice_pct )))\% & ((( sample.rna_metrics.pct_intergenic_bases|float2nice_pct )))\% \\
+        ((* if sample.rna_metrics.ribosomal_bases != "" *))
+        Ribosomal bases & ((( sample.rna_metrics.ribosomal_bases|nice_int ))) & ((( sample.rna_metrics.pct_ribosomal_bases_all|float2nice_pct )))\% & ((( sample.rna_metrics.pct_ribosomal_bases|float2nice_pct )))\% \\
+        ((* endif *))
+        \hline
+    \end{tabular}
+\end{center}