From a647f502c0d4d6527f8fe5a1904d319a7456777c Mon Sep 17 00:00:00 2001
From: Sigve Nakken <sigven@ifi.uio.no>
Date: Mon, 29 Jan 2024 11:16:48 +0100
Subject: [PATCH] fix alias type for cna/exp

---
 NEWS.md                        |   3 ++-
 R/sysdata.rda                  | Bin 751 -> 745 bytes
 data-raw/biomarker_utilities.R |  42 ++++++++++++++++++++++-----------
 3 files changed, 30 insertions(+), 15 deletions(-)

diff --git a/NEWS.md b/NEWS.md
index a21f58c..2b6eefd 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,7 +1,8 @@
-# Version 1.5.2 (January 28th 2024)
+# Version 1.5.3 (January 29th 2024)
 
 * Refine possible values in `treatment_category` argument to `get_drugs()` function
 * Fix anti-androgen classification
+* Fixed bug in alias type notation for copy numbers and expression biomarkers
 
 # Version 1.5.0 (January 25th 2024)
 
diff --git a/R/sysdata.rda b/R/sysdata.rda
index d07d020e84e319eadd3873bc3afe663f2a058b37..eec4a0bcfa8a75292a93bb6bae5df9f30de0ec5c 100644
GIT binary patch
literal 745
zcmV<F0v7#3T4*^jL0KkKS$(kc{{R8J|G@uy%5h)=f8am=|M0*6|L{Nn01?0fTpPPx
zvk=u86!t1@N2W}sn^QwasMAczq}rK0rbA5wK-0=*iM0rbdZrpjsk9;L851T<MgRaK
zAPku_&@vSx+M{JNQ_=yT&;S5v8UO$Q00AVEAv9tMf}WaW$^g@9G}F|`^iNPUVtSey
z2Z}f-nzO3N!YPR-0DagtySm2OamJ8-dLgp9RobV~-I6)mTnPkhC!JN1J+jlGL(1rQ
z=|2>8x?kLoL^wEKn>Rlr`*z*7i;RSCh!Y73k#WHjM94rLEH6VKATmM5E-`1(XcR!T
zYs4(#!*X$C!`V|l8LQ14kZ8AgMZzs&hbuq76!Z=?HEFwTzXuSirDl5eCp>^6{)c#|
z)t~^B8!)c(rz%KC6>Mr8qRTNFnHT_zMS!sd6Hne~V4)6H=rlXmz(Sm4bZ{PK$q6KO
zJWxgr6Pc00Qy7p14BJRD%Q#j}YG0ZtI=5v~TT<4~N!g~OdrIi4crB5uGBJ94Xt~Iy
znehm?{Hd^$HPsyFNwuW(Jlbnq*2LLR0)DHb3Y?{{(`%kle;n^N%xy0f+Z+)wQA~vb
zx*#ZkBoi2KkYy5HGcc-7JP`2}#;w`cYl83h=df!+^$424<A|8>-!B)6G3azObIId=
zTO;9t^t7!`IURL)@z5j@D#lR|Y-7DxL7+SAnA4I~Brxxi$C8KwIe%wOIko83t6xA#
zwbK?;Bi1+B8Hnjf?oT{8`4=(+>~6nKf$VYEI$FLPsTHT`s!nt4dmEYXW(f+SZ>Vl=
zEON$aD~{0Etq_Ih7zptdJP*k(I27e;S;^|%v!`~zjV8*8IO%MBk(j-dl;7JDH!|X&
zMBCf=sk<EXidc+tIQ7a(&k)<~2&iN^!wlX>^;vW3jfIU)IxKgph@MeU-YqWn-|ODf
blBQy$dzoR=vUG?*Pw{soQ-ui!+YdkR5A0hx

literal 751
zcmV<L0ucQ|T4*^jL0KkKS-%M7kpKa_f588H%5h)=f8aj<|M0*6|KLCX01?0fTpOL-
zX6!X7>7+FeN$IB4Wi!<bX&XaL5sghWGG>4iOcCT7gr}7DRPr{NH8jZ7+G+ze0i)6k
zfM@_{$Y@12ff{NwYI>OUG(!{0G%y+f5s9Dx&=^T1(W-iy9+S}284VKz@}8QWlh6nO
z=xQ@lOwihn4J`c!RD%ebB%A^MV5Q#XDVu6Uga3*=)l#a`_B!+(l9vKO1{2JBDAJ&=
z=;nE*hmMDY-^XWVglLV|$#1_DeE(i4(OHm-Vg$lMWL9uQ(J~MRTMEM<BN-s$76E5>
zVFeH`@2sNc^5qq(BNpbhtemo0o?jM&+1t6S3iz`M5SgQ%zJ{68Cg!0M@L|4U*<efn
zmLDNB@raWAfQjjM!5R;w>ep){Ka&}<QqUsgfV1!#288cCmH;#_sv(HTN*&-8!EA+;
z$W1xB127F831)N@Kxoi5@Jb%@gXx;${|tkxh6h|qR9T6KyZVQ(6qY=>Ss_{Mg1*d@
zelN976&vE?KCDlBbEtX2Wa;yvUg4(iLpb58sg^uXnOE1YX&0zHwX03aD(tX31COAF
z9>r^Sij`&HduraA?Yyy1h7?jZh`Ip-j{^&VgK>dDTlGraP8zs3HzKSZsxsb(oXS<y
zBGOk{W03BZ!o=^Br|O$*Y6ZtD&!Od%)8j;a&8Ay1Tps^FWiQ5jxwB`y2*-*nl-m?Z
z4c%j3nYpUjo{GmsDT0`N27`<O{5wI4O?0NW=B6eo#jj+Wz=4sw8nJceY77VuiNbyG
ze<Fkk{vma<tdPqgFPTgvk8yPEeoF)ygOJG~4q-A}0e4fWgRnP-U_XQsolO>E$#bs2
zjw`rv{j2YB;+JB9YD8UQ@hK}jMDDoKKD!}hC>%gZ{~sf1A=Vd0xd4Dv5x{5^AkxHj
h!&jypRp6e!J4e$QLh*-@3Q%5x?ntK!5*OhdvLLdsV3_~_

diff --git a/data-raw/biomarker_utilities.R b/data-raw/biomarker_utilities.R
index 84774f4..dc41d76 100644
--- a/data-raw/biomarker_utilities.R
+++ b/data-raw/biomarker_utilities.R
@@ -1460,7 +1460,12 @@ load_civic_biomarkers <- function(
         relationship = "many-to-many"
       ) |>
       dplyr::distinct()
-  )
+  ) |>
+    dplyr::mutate(alias_type = dplyr::if_else(
+      alias_type == "other",
+      "other_gene",
+      as.character(alias_type)
+    ))
 
   return(biomarker_items)
 
@@ -1598,7 +1603,8 @@ load_cgi_biomarkers <- function(compound_synonyms = NULL,
     )) |>
     tidyr::separate(gene, c("gene1","gene2"), sep = ";") |>
     dplyr::mutate(gene = dplyr::if_else(
-      !is.na(gene1) & !is.na(gene2) & startsWith(molecular_profile, gene2),
+      !is.na(gene1) & !is.na(gene2) & 
+        startsWith(molecular_profile, gene2),
       as.character(gene2),
       as.character(gene1)
     )) |>
@@ -1868,20 +1874,20 @@ load_cgi_biomarkers <- function(compound_synonyms = NULL,
     ## get variant id's per molecular profile
     dplyr::inner_join(
       dplyr::select(
-        cgi_variants, entrezgene, variant_id, 
-        variant_consequence, variant_alias,
-        alteration_type),
-      multiple = "all", relationship = "many-to-many"
+        cgi_variants, variant_id, 
+        gene, alias_type, variant_alias),
+      by = c("gene", "variant_alias"),
+      relationship = "many-to-many"
     ) |>
     
     ## add all variant aliases
-    dplyr::select(-variant_alias) |>
-    dplyr::distinct() |>
-    dplyr::left_join(
-      dplyr::select(cgi_variants, entrezgene, variant_id, 
-                    variant_consequence, variant_alias, alias_type,
-                    alteration_type),
-      multiple = "all", relationship = "many-to-many")  |>
+    # dplyr::select(-variant_alias) |>
+    # dplyr::distinct() |>
+    # dplyr::left_join(
+    #   dplyr::select(cgi_variants, entrezgene, variant_id, 
+    #                 variant_consequence, variant_alias, alias_type,
+    #                 alteration_type),
+    #   relationship = "many-to-many")  |>
     dplyr::select(-c(entrezgene, variant_consequence, variant_alias,
                      alias_type, gene, symbol, alteration_type,
                      alteration)) |>
@@ -1987,9 +1993,17 @@ load_cgi_biomarkers <- function(compound_synonyms = NULL,
       dplyr::select(
         biomarker_items$clinical,
         variant_id
-      ), relationship = "many-to-many"
+      ), by = "variant_id", 
+      relationship = "many-to-many"
     ) |>
      dplyr::distinct() 
+  ) |>
+  dplyr::mutate(
+    alias_type = dplyr::case_when(
+      stringr::str_detect(alteration_type,"^EXP") |
+        alias_type == "other" ~ "other_gene",
+      TRUE ~ as.character(alias_type)
+    )
   )
 
   return(biomarker_items)