diff --git a/NEWS.md b/NEWS.md
index f41fb8f307..953c54894b 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -14,6 +14,12 @@
   when a vector of duplicate indices is passed when doing column selection
   ([#3302](https://github.com/JuliaData/DataFrames.jl/pull/3302))
 
+## Bug fixes
+
+* `deleteat!` correctly handles the situation when vector of rows to be dropped
+  from a data frame is its column or might alias with some of its columns
+  ([#3304](https://github.com/JuliaData/DataFrames.jl/pull/3304))
+
 # DataFrames.jl v1.5 Release Notes
 
 ## New functionalities
diff --git a/src/dataframe/dataframe.jl b/src/dataframe/dataframe.jl
index 5e59a2d068..0c63fb14de 100755
--- a/src/dataframe/dataframe.jl
+++ b/src/dataframe/dataframe.jl
@@ -904,6 +904,10 @@ function _deleteat!_helper(df::DataFrame, drop)
         return df
     end
 
+    if any(c -> c === drop || Base.mightalias(c, drop), cols)
+        drop = copy(drop)
+    end
+
     n = nrow(df)
     col1 = cols[1]
     deleteat!(col1, drop)
diff --git a/test/dataframe.jl b/test/dataframe.jl
index d9cce76828..0937e214a2 100644
--- a/test/dataframe.jl
+++ b/test/dataframe.jl
@@ -518,6 +518,13 @@ end
     df = DataFrame(a=1, b=3.0)
     @test isempty(deleteat!(df, true:true))
 
+    df = DataFrame(a=[false, true, true], b=1:3, c=4:6)
+    @test deleteat!(df, df.a) == DataFrame(a=false, b=1, c=4)
+    df = DataFrame(a=1:3, b=[false, true, true], c=4:6)
+    @test deleteat!(df, df.b) == DataFrame(a=1, b=false, c=4)
+    df = DataFrame(a=1:3, b=4:6, c=[false, true, true])
+    @test deleteat!(df, df.c) == DataFrame(a=1, b=4, c=false)
+
     Random.seed!(1234)
     for t in 0:0.005:1.0
         # two columns are needed as the second column is affected