Tp53 loss status update: bug fix (#942)

* ficing bug

* removing incorrect commited files

* summarize to max within 1 and 0

* Update README.md

* resize plots

analyses/tp53_nf1_score/04-tp53-altered-annotation.Rmd

@@ -183,8 +183,7 @@ tp53_alterations_dna <- histology %>%
   # join filtered cnv losses
   left_join(cnv_domain_overlap,by=c("Kids_First_Biospecimen_ID"="biospecimen_id",
                                     "sample_id")) %>%
-  rename(Kids_First_Biospecimen_ID_DNA = Kids_First_Biospecimen_ID) %>%
-  replace_na(list(hotspot = 0, activating = 0))
+  rename(Kids_First_Biospecimen_ID_DNA = Kids_First_Biospecimen_ID) 
 
 tp53_alterations_score <- tp53_alterations_dna %>%
   # add classifier score
@@ -197,7 +196,8 @@ tp53_alterations_score <- tp53_alterations_dna %>%
   select(sample_id,Kids_First_Biospecimen_ID_RNA,Kids_First_Biospecimen_ID_DNA,
          tp53_score,cancer_predispositions,HGVSp_Short,copy_number,hotspot,activating) %>%
   arrange(sample_id) %>%
-  unique() 
+  unique() %>%
+  replace_na(list(hotspot = 0, activating = 0))
 
 tp53_alterations_score 
 
@@ -224,8 +224,8 @@ tp53_alterations_score_wide <- tp53_alterations_score %>%
     HGVSp_Short = toString(unique(HGVSp_Short)),
     CNV_loss_evidence = toString(unique(copy_number)),
     # summarize unique hotspot values per sample_id
-    hotspot = toString(unique(hotspot[!is.na(hotspot) ])),
-    activating = toString(unique(activating[!is.na(activating)]))) 
+    hotspot = max(unique(hotspot[!is.na(hotspot) ])),
+    activating = max(unique(activating[!is.na(activating)]))) 
 
 tp53_alterations_score_wide
 
@@ -242,16 +242,16 @@ tp53_alterations_score_wide <- tp53_alterations_score_wide %>%
     # add tp53_altered annotation column
     tp53_altered = 
       case_when(
-        # when activating == ""
-        activating == "" &
+        # when activating == 0
+        activating == 0 &
           # check if mutated variant AA position overlaps hotspot database 
           ( hotspot == 1  |
               # check if sample_id has SNV+CNV mutation
               # suggesting both alleles are mutated
               (SNV_indel_counts >= 1 & CNV_loss_counts >=1) |
               # check if more than 1 SNV is present
               # suggesting both alleles are mutated
-              SNV_indel_counts >= 1 |
+              SNV_indel_counts > 1 |
               # check if SNV mutant and has
               # Li-Fraumeni syndrome suggesting
               # germline TP53 mutant as cancer predisposition 
@@ -283,7 +283,7 @@ tp53_alterations_score_wide <- tp53_alterations_score_wide %>%
 
 ## Explore distribution of tp53_altered status vs tp53 inactivation scores
 
-```{r}
+```{r, out.width="50%"}
 ggplot(tp53_alterations_score_wide, aes(x = factor(tp53_altered), y = tp53_score)) +
   geom_violin()+
   geom_jitter(alpha = 0.5, width = 0.2) +
@@ -318,7 +318,7 @@ stranded_tp53 <- as.data.frame(subset_stranded) %>%
 
 Plot distribution of TP53 gene expression (stranded)
 
-```{r}
+```{r, out.width="50%"}
 
 ggplot(stranded_tp53, aes(x = factor(tp53_altered), y = TP53)) +
   geom_violin()+
@@ -350,7 +350,7 @@ polya_tp53<- as.data.frame(subset_polya) %>%
 
 Plot distribution of TP53 gene expression (poly-a)
 
-```{r}
+```{r, out.width="50%" }
 
 ggplot(polya_tp53, aes(x = factor(tp53_altered), y = TP53)) +
   geom_violin()+
@@ -365,7 +365,7 @@ ggplot(polya_tp53, aes(x = factor(tp53_altered), y = TP53)) +
 
 ### Check if other cancer predisposition have high TP53 inactivation scores  
 
-```{r}
+```{r ,out.width="50%"}
 
 ggplot(tp53_alterations_score_wide, aes(x = factor(tp53_altered), y = tp53_score)) +
   geom_violin()+

analyses/tp53_nf1_score/README.md

@@ -47,7 +47,8 @@ SNV_indel_counts	| Number of deleterious SNVs found in DNA sample
 CNV_loss_counts	| Number of CNV losses found in DNA sample
 HGVSp_Short	| Short format of protein level change used as SNV evidence 
 CNV_loss_evidence | copy_number of CNV overlapping functional domains of TP53 used as evidence 	
-hotspot	activating	| Hotspot status from MSKCC cancer hotspot [database](https://www.cancerhotspots.org/#/home) 
+hotspot	| Any 1 SNV shown in HGVSp_Short overlaps MSKCC cancer hotspot [database](https://www.cancerhotspots.org/#/home)
+activating	| Any 1 SNV shown in HGVSp_Short overlaps TP53_ activating mutations R273C and R248W. [Reference](https://pubmed.ncbi.nlm.nih.gov/17417627/) and [reference](https://pubmed.ncbi.nlm.nih.gov/24677579/). 
 tp53_altered | Combined evidence, cancer predisposition and score based tp53 status