Merge pull request #27 from SorenKarst/develop

Fix racon bug and integrate UMI filtering in pipeline
SorenKarst · Feb 9, 2020 · ccfd11f · ccfd11f
2 parents 9ff8860 + 327c79f
commit ccfd11f
Show file tree

Hide file tree

Showing 6 changed files with 108 additions and 108 deletions.
diff --git a/README.md b/README.md
@@ -206,7 +206,7 @@ Tools:
 For help with a specific tool or pipeline:
 longread_umi <name> -h
 ```
-
+  
 
 
 ```
@@ -227,7 +227,7 @@ where:
     -n  Process n number of bins. If not defined all bins
         are processed.
 ```
-
+  
 
 
 ```
@@ -257,7 +257,7 @@ where:
     -n  Barcode numbers used. [Default  = '1-120'].
     -t  Number of threads used.
 ```
-
+  
 
 
 ```
@@ -292,7 +292,7 @@ where:
         sequences. Default 2.
     -t  Number of threads used.
 ```
-
+  
 
 
 ```
@@ -331,7 +331,7 @@ where:
     -T  Number of medaka jobs to start. Threads pr. job is threads/jobs.
         [Default = 1].
 ```
-
+  
 
 
 ```
@@ -379,7 +379,7 @@ longread_umi nanopore_settings_test
   -y 3 
   -n 1000
 ```
-
+  
 
 
 ```
@@ -413,7 +413,7 @@ where:
     -u  Directory with UMI binned reads.
     -t  Number of threads to use.
 ```
-
+  
 
 
 ```
@@ -436,7 +436,7 @@ where:
         are processed.
     -t  Number of Medaka jobs to run. [Default = 1].
 ```
-
+  
 
 
 ```
@@ -461,7 +461,7 @@ where:
     -e  Length of terminal end to search for primers. [Default = 500]
     -n  Subset reads before search. [Default = 100000]
 ```
-
+  
 
 
 ```
@@ -503,7 +503,7 @@ wget https://www.arb-silva.de/fileadmin/silva_databases/
 release_132/Exports/SILVA_132_SSURef_Nr99_tax_silva.fasta.gz
 gunzip SILVA_132_SSURef_Nr99_tax_silva.fasta.gz
 ```
-
+  
 
 
 ```
@@ -527,7 +527,7 @@ where:
     -t  Number of threads to use.
     -l  Log directory
 ```
-
+  
 
 
 ```
@@ -537,7 +537,8 @@ where:
    adaptor regions.
 
 usage: umi_binning [-h] (-d file -o dir -m value -M value )
-(-s value -e value -f string -F string -r string -R string -p -t value) 
+(-s value -e value -f string -F string -r string -R string -p )
+(-u value -U value -O value -S value -t value) 
 
 where:
     -h  Show this help text.
@@ -551,10 +552,18 @@ where:
     -F  Forward primer sequence.
     -r  Reverse adaptor sequence.
     -R  Reverse primer sequence.
-    -p  Flag to disable Nanopore trimming and filtering. Use with PacBio reads.
+    -p  Flag to disable Nanopore trimming and filtering.
+        Use with PacBio reads.
+    -u  Discard bins with a mean UMI match error above u.
+    -U  Discard bins with a UMI match error standard
+        deviation above U.
+    -O  Normalize read orientation fraction to 'O' if < 'O' reads are
+        either +/- strand orientation.
+    -N  Max number of reads with +/- orientation. [Default = 10000]
+    -S  UMI bin size/UMI cluster size cutoff. [Default = 10]
     -t  Number of threads to use.
 ```
-
+  
 
 
 ```
@@ -576,7 +585,7 @@ where:
     -t  Number of threads to use. [Default = 1]
     -b  Debug flag. Keep temp files. [Default = NO]
 ```
-
+  
 
 
 

diff --git a/docs/ONT_R10_ZYMO_rRNA.html b/docs/ONT_R10_ZYMO_rRNA.html
diff --git a/docs/ONT_R10_ZYMO_rRNA.rmd b/docs/ONT_R10_ZYMO_rRNA.rmd
@@ -19,7 +19,7 @@
 
 * Flowcell: R10
 * Instrument: MinION
-* Basecalling: guppy v3.4.4
+* Basecalling: guppy v3.4.4 with HAC model
 
 ### Generate UMI consensus sequences
 
@@ -46,7 +46,7 @@ longread_umi nanopore_pipeline \
   -R CGACATCGAGGTGCCAAAC \
   -c 2 \
   -p 2 \
-  -t 40 \
+  -t 10 \
   -T 1
 ```
 
@@ -58,7 +58,7 @@ longread_umi qc_pipeline \
 -r "zymo_curated" \
 -u umi_out \
 -o umi_out/qc \
--t 40
+-t 10
 ```
 
 ## Validate UMI consensus sequences
@@ -97,15 +97,11 @@ load("validation/ONT_R10_ZYMO_rRNA_ep.Rdata")
 ```
 
 
-Perform UMI bin post-processing filtering (will be incoporated in pipeline in the future)
+Remove UMI sequences without primers in both ends
 ```{r echo=TRUE, message=FALSE, warning=FALSE}
 qc <- filter(
   qc,
-  ror > 10^-0.6, # remove bins with extreme read orientation ratios
-  ror < 10^0.6,
-  umi_bin_size/umi_cluster_size < 10, # remove bins with extreme bin to cluster size
-  umi_match_error < 3.5, # remove bins with high mismatch between UMI reference and reads
-  !is.na(length) # remove umi bins where >=1 gene specific primers was not found in consensus sequence.
+  !is.na(length) # remove umi bins where one or both gene specific primers was not found in the consensus sequence.
 )
 
 ep <- filter(
@@ -170,38 +166,38 @@ Flag contamination and chimeras. Flagged chimeras and contamination below match
 ```{r echo=TRUE, message=FALSE, warning=FALSE}
 cont <- tibble(
   umi = c(
-    "umi47838bins",
-    "umi177241bins",
-    "umi13630bins",
-    "umi173605bins",
-    "umi159967bins",
-    "umi185039bins",
-    "umi181256bins",
-    "umi6705bins",
-    "umi42968bins",
-    "umi238547bins",
-    "umi199820bins",
-    "umi151265bins",
-    "umi6094bins",
-    "umi193827bins",
-    "umi181472bins",
-    "umi198347bins",
-    "umi234765bins",
-    "umi180359bins",
-    "umi180080bins",
-    "umi17145bins",
-    "umi189949bins"
-),
+    "umi191103bins",
+    "umi111976bins",
+    "umi217947bins",
+    "umi185324bins",
+    "umi191976bins",
+    "umi44490bins",
+    "umi14613bins",
+    "umi181378bins",
+    "umi6637bins",
+    "umi151082bins",
+    "umi35088bins",
+    "umi173219bins",
+    "umi163928bins",
+    "umi181158bins",
+    "umi6090bins",
+    "umi233806bins",
+    "umi19568bins",
+    "umi187615bins",
+    "umi185965bins",
+    "umi171705bins",
+    "umi171734bins"
+  ),
   flag = "contamination"
 )
 
 chi <- tibble(
   umi = c(
-    "umi176051bins",
-    "umi12507bins",
-    "umi183450bins",
-    "umi146629bins"
-),
+    "umi12455bins",
+    "umi178211bins",
+    "umi158224bins",
+    "umi176163bins"
+  ),
   flag = "chimera"
 )
 ```

diff --git a/docs/rdata/ONT_R10_ZYMO_rRNA_ep.Rdata b/docs/rdata/ONT_R10_ZYMO_rRNA_ep.Rdata
diff --git a/docs/rdata/ONT_R10_ZYMO_rRNA_qc.Rdata b/docs/rdata/ONT_R10_ZYMO_rRNA_qc.Rdata
diff --git a/scripts/validation_functions.R b/scripts/validation_functions.R
@@ -161,8 +161,8 @@ lu_compile_qc <- function(
     dplyr::transmute(umi = gsub(";.*", "", umi),
                      ref_error = error,
                      ref_tax = target)
-
-  if(!is.null(ref_ssu_sam)){  
+  
+  if(!is.null(silva)){
     ref_ssu_error <- readr::read_delim(
       file = paste(data_dir, "/", ref_ssu_sam, sep = ""),
       delim = "\t",
@@ -171,9 +171,7 @@ lu_compile_qc <- function(
       dplyr::transmute(umi = gsub(";.*", "", umi),
                        ref_ssu_error = error,
                        ref_ssu_tax = target)
-  }
-
-  if(!is.null(silva)){
+
     silva_error <- readr::read_delim(
       file = paste(data_dir, "/", silva_ssu_sam, sep = ""),
       delim = "\t",
@@ -248,7 +246,7 @@ lu_compile_qc <- function(
   qc <-  umi_stats %>%
     left_join(con_length, by = "umi") %>%
     left_join(ref_error, by = "umi") %>%
-    {if(!is.null(ref_ssu_sam))left_join(., ref_ssu_error, by = "umi") else .} %>%
+    {if(!is.null(silva))left_join(., ref_ssu_error, by = "umi") else .} %>%
     left_join(chimera, by = "umi") %>%
     {if(!is.null(silva))left_join(., silva, by = "umi") else .} %>%
     {if(!is.null(read_orientation))left_join(., ror, by = "umi") else .}
@@ -1103,6 +1101,7 @@ lu_errortype_plot <- function(
 lu_errortype_plot_tbl <- function(
                               profile,
                               break_size = 5,
+                              digits = 3,
                               title){
 
   # Summarise based on error type
@@ -1143,7 +1142,7 @@ lu_errortype_plot_tbl <- function(
     # Format decimals
     mutate_at(
       vars(all_err:del_err),
-      ~round(., 3)
+      ~round(., digits)
     )
     colnames(errortype_tbl) <- c(
       "UMI bin size",