new color scheme mode in alignedSeq

CaoTianze · Aug 24, 2023 · fce95f9 · fce95f9
1 parent 6e26fd0
commit fce95f9
Show file tree

Hide file tree

Showing 12 changed files with 138 additions and 48 deletions.
diff --git a/docs/geom_alignedSeq.md b/docs/geom_alignedSeq.md
@@ -1,12 +1,30 @@
 # geom_alignedSeq
 A class that represents the sequence alignment diagram
-## *class* geom_alignedSeq(self,data: Union[list[str], dict, None] = None,seq_names: Optional[list[str]] = None,seq_type: str = 'AUTO',namespace: Optional[list[str]] = None,font: str = 'roboto_medium',stack_width: float = 0.75,font_col: str = '#FFFFFF',bg_col_scheme: Union[DataFrame, str] = 'AUTO',bg_low_col: str = 'black',bg_high_col: str = 'yellow',bg_na_col: str = '#333333',**kwargs: Any)
+## *Sample code*
 ```python
 from plotnine import ggplot, coord_fixed
 from plotnineseqsuite import geom_alignedSeq, theme_seq
 from plotnineseqsuite.data import seqs_dna
 ggplot() + geom_alignedSeq(seqs_dna['MA0013.1']) + theme_seq() + coord_fixed()
 ```
+## *Init param*
+```
+geom_alignedSeq(self,
+             data: list[str] | dict | None = None,
+             seq_names: list[str] | None = None,
+             seq_type: str = 'AUTO',
+             namespace: list[str] | None = None,
+             font: str = 'roboto_medium',
+             stack_width: float = 0.75,
+             border_col: str = 'grey',
+             scheme_applied: str = 'BACKGROUND',
+             no_scheme_col: str = '#000000',
+             col_scheme: DataFrame | str = 'AUTO',
+             low_col: str = 'black',
+             high_col: str = 'yellow',
+             na_col: str = '#333333',
+             **kwargs: Any) -> Any
+```
 - data    
 Sequence data or corresponding dict.
 - seq_names    
@@ -19,16 +37,20 @@ The letter corresponding to the data.
 Font value
 - stack_width    
 The ratio of the size of letters to the standard unit width.
-- font_col    
-The color of the font.
-- bg_col_scheme    
-Color scheme of the backgrounds.
-- bg_low_col    
+- border_col
+The color of the border in the background. When it is None, the border of the background will disappear.
+- scheme_applied
+BACKGROUND or LETTER. Indicates the target to which the color scheme applies.
+- no_scheme_col    
+When a color scheme is applied to the background, this indicates the color of the letters. When a color scheme is applied to letters, this indicates the color of the background.
+- col_scheme    
+Color scheme.
+- low_col    
 Continuous color schemes are available.
-- bg_high_col    
+- high_col    
 Continuous color schemes are available.
-- bg_na_col    
-Used when the background in the corresponding namespace do not have a color matching value defined.
+- na_col    
+It is available when the corresponding namespace do not have a color matching value defined.
 - kwargs    
 Other arguments passed on to layer().
 ## *properties*

diff --git a/docs/geom_logo.md b/docs/geom_logo.md
@@ -1,12 +1,28 @@
 # geom_logo
 A class that represents the sequence logo
-## *class* geom_logo(self,data: Union[list[str], ndarray, dict, None] = None,method: str = 'bits',seq_type: str = 'AUTO',namespace: Optional[list[str]] = None,font: str = 'roboto_medium',stack_width: float = 0.95,rev_stack_order: bool = False,col_scheme: Union[DataFrame, str] = 'AUTO',low_col: str = 'black',high_col: str = 'yellow',na_col: str = '#333333',**kwargs: Any)
+## *Sample code*
 ```python
 from plotnine import ggplot
 from plotnineseqsuite import geom_logo, theme_seq
 from plotnineseqsuite.data import seqs_dna
 ggplot() + geom_logo(seqs_dna['MA0001.1']) + theme_seq()
 ```
+## *Init param*
+```
+geom_logo(self,
+             data: list[str] | ndarray | dict | None = None,
+             method: str = 'bits',
+             seq_type: str = 'AUTO',
+             namespace: list[str] | None = None,
+             font: str = 'roboto_medium',
+             stack_width: float = 0.95,
+             rev_stack_order: bool = False,
+             col_scheme: DataFrame | str = 'AUTO',
+             low_col: str = 'black',
+             high_col: str = 'yellow',
+             na_col: str = '#333333',
+             **kwargs: Any) -> Any
+```
 - data    
 Sequence data or PFM or corresponding dict.
 - method    

diff --git a/docs/geom_seqBar.md b/docs/geom_seqBar.md
@@ -1,12 +1,27 @@
 # geom_seqBar
 A class that represents the sequence histogram
-## *class* geom_seqBar(self,data: Union[list[str], ndarray, dict, None] = None,seq_type: str = 'AUTO',namespace: Optional[list[str]] = None,font: str = 'roboto_medium',stack_width: float = 0.75,bar_col_scheme: Union[DataFrame, str] = 'AUTO',font_col: str = '#808080',low_col: str = 'black',high_col: str = 'yellow',na_col: str = '#333333',**kwargs: Any)
+## *Sample code*
 ```python
 from plotnine import ggplot
 from plotnineseqsuite import geom_seqBar, theme_seq
 from plotnineseqsuite.data import seqs_dna
 ggplot() + geom_seqBar(seqs_dna['MA0013.1']) + theme_seq()
 ```
+## *Init param*
+```
+geom_seqBar(self,
+             data: list[str] | ndarray | dict | None = None,
+             seq_type: str = 'AUTO',
+             namespace: list[str] | None = None,
+             font: str = 'roboto_medium',
+             stack_width: float = 0.75,
+             bar_col_scheme: DataFrame | str = 'AUTO',
+             font_col: str = '#808080',
+             low_col: str = 'black',
+             high_col: str = 'yellow',
+             na_col: str = '#333333',
+             **kwargs: Any) -> Any
+```
 - data    
 Sequence data or PFM or corresponding dict.
 - seq_type    

diff --git a/docs/index.md b/docs/index.md
@@ -26,17 +26,25 @@ geom_alignedSeq accepts two types of input, each described in detail below
 1. list: a list of aligned sequences
 2. dict: It is used for plotting more than one sequence logo at the same time with the help of facets, the key of dict is facet value, and the value of dict is list described above
 ### No background color
-When bg_col_scheme option is None, the picture has no background color. 
+When col_scheme option is None, the picture has no background color. 
 ```python
-ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], font_col='black', bg_col_scheme=None) + theme_seq() + coord_fixed()
+ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], no_scheme_col='black', col_scheme=None) + theme_seq() + coord_fixed()
 ```
 ![](no_background_color.png)
 ### No sequence letter
-When font option is None, the picture has only the background color. 
+When font option is None and border_col is None, the picture has only the background color. 
 ```python
-ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], font=None) + theme_seq() + coord_fixed()
+ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], font=None, border_col=None) + theme_seq() + coord_fixed()
 ```
 ![](no_sequence_letter.png)
+
+### The color scheme switches to letters
+When the option scheme_applied is LETTER, the color scheme will be applied from the background to the letters.
+```Python
+ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], no_scheme_col='white', scheme_applied='LETTER') + theme_seq() + coord_fixed()
+```
+![](switch_color_schme.png)
+
 ### Tagging sequences
 You can set seq_names parameter to identify the name of the sequence.
 ```python
@@ -133,7 +141,7 @@ ggplot() + geom_logo(seqs_numeric, method='probability', namespace=['δ','ε','
 ![](custom_alphabet_2.png)
 ## Colour schemes
 ### Preset color schemes
-plotnineSeqSuite has preset color schemes that can be set using the col_scheme parameter in geom_logo, the parameter bar_col_scheme in geom_seqBar and the parameter bg_col_scheme in geom_alignedSeq. By default, the col_scheme is set to AUTO such that the color scheme is automatically chosen based on your sequence type.
+plotnineSeqSuite has preset color schemes that can be set using the col_scheme parameter in geom_logo, the parameter bar_col_scheme in geom_seqBar and the parameter col_scheme in geom_alignedSeq. By default, the col_scheme is set to AUTO such that the color scheme is automatically chosen based on your sequence type.
 
 Lets try generate an amino acid sequence logo using kinase-substrate phosphorylation data:
 ```python

diff --git a/docs/index.pdf b/docs/index.pdf
diff --git a/docs/no_sequence_letter.png b/docs/no_sequence_letter.png
diff --git a/docs/plot_an_aligned_sequences.png b/docs/plot_an_aligned_sequences.png
diff --git a/docs/switch_color_schme.png b/docs/switch_color_schme.png
diff --git a/plotnineseqsuite/align.py b/plotnineseqsuite/align.py
@@ -17,12 +17,14 @@ class geom_alignedSeq:
     def __init__(self, data: Union[List[str], Dict] = None, seq_names: List[str] = None, seq_type: str = 'AUTO',
                  namespace: List[str] = None,
                  font: str = 'roboto_medium', stack_width: float = 0.75, border_col: str = 'grey',
-                 font_col: str = '#000000', bg_col_scheme: Union[DataFrame, str] = 'AUTO',
-                 bg_low_col: str = 'black', bg_high_col: str = 'yellow', bg_na_col: str = '#333333',
+                 scheme_applied: str = 'BACKGROUND', no_scheme_col: str = '#000000',
+                 col_scheme: Union[DataFrame, str] = 'AUTO',
+                 low_col: str = 'black', high_col: str = 'yellow', na_col: str = '#333333',
                  **kwargs):
         self.__kwargs = kwargs
-        self.__font_col = font_col
+        self.__no_scheme_col = no_scheme_col
         self.__border_col = border_col
+        self.__scheme_applied = scheme_applied
         if stack_width > 1 or stack_width <= 0:
             raise Exception('"stack_width" must be between 0 and 1')
         if data is None:
@@ -31,6 +33,8 @@ def __init__(self, data: Union[List[str], Dict] = None, seq_names: List[str] = N
             seq_type = 'OTHER'
         if seq_type not in {'OTHER', 'AUTO', 'AA', 'DNA', 'RNA'}:
             raise Exception("seq_type must be one of 'OTHER' or 'AUTO', or 'AA', or 'DNA', or 'RNA'")
+        if scheme_applied not in {'BACKGROUND', 'LETTER'}:
+            raise Exception("scheme_applied must be one of 'BACKGROUND' or 'LETTER'")
         if type(data) == list:
             data = {1: data}
         lvls = data.keys()
@@ -43,40 +47,48 @@ def __init__(self, data: Union[List[str], Dict] = None, seq_names: List[str] = N
         if font is not None:
             letter_data = reduce(lambda x, y: concat([x, y]), map(lambda x: x['letter_data'], data_sp))
 
-        if type(bg_col_scheme) is str:
-            bg_cs_dict = get_col_scheme(bg_col_scheme, seq_type)
-        elif type(bg_col_scheme) is dict and type(bg_col_scheme['cs']) == DataFrame:
-            bg_cs_dict = bg_col_scheme
+        if type(col_scheme) is str:
+            cs_dict = get_col_scheme(col_scheme, seq_type)
+        elif type(col_scheme) is dict and type(col_scheme['cs']) == DataFrame:
+            cs_dict = col_scheme
 
-        if bg_col_scheme is not None:
-            legend_title = bg_cs_dict['name']
-            colscale_gradient = True if is_numeric_dtype(bg_cs_dict['cs']['group']) else False
+        if col_scheme is not None:
+            legend_title = cs_dict['name']
+            colscale_gradient = True if is_numeric_dtype(cs_dict['cs']['group']) else False
             if colscale_gradient:
-                bg_cs_dict['cs'] = concat([bg_cs_dict['cs'],DataFrame(data={'letter':['-'],'group':[0]})])
-                colscale_opts = scale_fill_gradient(low=bg_low_col, high=bg_high_col, name=legend_title,
-                                                    na_value=bg_na_col)
+                # TODO check contain -, then add -
+                cs_dict['cs'] = concat([cs_dict['cs'], DataFrame(data={'letter': ['-'], 'group': [0]})])
+                colscale_opts = scale_fill_gradient(low=low_col, high=high_col, name=legend_title,
+                                                    na_value=na_col)
             else:
-                bg_cs_dict['cs'] = concat([bg_cs_dict['cs'], DataFrame(data={'letter': ['-'], 'group': ['-'], 'col': ['#FFFFFF']})])
-                tmp = bg_cs_dict['cs'].drop_duplicates(subset=['group']).dropna(subset=['group'])
+                cs_dict['cs'] = concat(
+                    [cs_dict['cs'], DataFrame(data={'letter': ['-'], 'group': ['-'], 'col': ['#FFFFFF']})])
+                tmp = cs_dict['cs'].drop_duplicates(subset=['group']).dropna(subset=['group'])
                 col_map = {}
                 for item in map(lambda x, y: {x: y}, tmp['group'], tmp['col']):
                     col_map.update(item)
-                colscale_opts = scale_fill_manual(values=col_map, name=legend_title, na_value=bg_na_col)
-            bg_data = merge(bg_data, bg_cs_dict['cs'], how='left')
-            self.bg_data = bg_data
+                colscale_opts = scale_fill_manual(values=col_map, name=legend_title, na_value=na_col)
+            if scheme_applied == 'BACKGROUND':
+                bg_data = merge(bg_data, cs_dict['cs'], how='left')
+            else:
+                letter_data = merge(letter_data, cs_dict['cs'], how='left')
             self.colscale_opts = colscale_opts
         else:
-            self.bg_data = None
+            if scheme_applied == 'BACKGROUND':
+                bg_data = None
+            else:
+                letter_data = None
             self.colscale_opts = None
 
         if font is not None:
             letter_data = letter_data.sort_values(by='order').reset_index(drop=True)
             letter_data['group_by'] = letter_data.apply(
                 lambda x: '{}.{}.{}.{}'.format(x['seq_group'], x['letter'], x['position'], x['y_index']), axis=1)
-            self.letter_data = letter_data
         else:
-            self.letter_data = None
-        self.scale_x_continuous = scale_x_continuous(breaks=lambda x: range(floor(x[0]), ceil(x[1])), expand=(0,0))
+            letter_data = None
+        self.bg_data = bg_data
+        self.letter_data = letter_data
+        self.scale_x_continuous = scale_x_continuous(breaks=lambda x: range(floor(x[0]), ceil(x[1])), expand=(0, 0))
         self.scale_y_continuous = scale_y_continuous(breaks=None)
         if seq_names is not None:
             self.scale_y_continuous = scale_y_continuous(breaks=lambda x: [k + 0.5 for k in range(0, int(x[1]))],
@@ -130,15 +142,27 @@ def to_df_data(x, y_index):
     def __radd__(self, gg):
         params = []
         if self.bg_data is not None:
-            bg_layer = geom_tile(data=self.bg_data,
-                                 mapping=aes(x='x', y='y', width='width', height='height', fill='group'),
-                                 color=self.__border_col,
-                                 **self.__kwargs)
+            if self.__scheme_applied == 'BACKGROUND':
+                bg_layer = geom_tile(data=self.bg_data,
+                                     mapping=aes(x='x', y='y', width='width', height='height', fill='group'),
+                                     color=self.__border_col,
+                                     **self.__kwargs)
+            else:
+                bg_layer = geom_tile(data=self.bg_data,
+                                     mapping=aes(x='x', y='y', width='width', height='height'),
+                                     fill=self.__no_scheme_col,
+                                     color=self.__border_col,
+                                     **self.__kwargs)
             params.append(bg_layer)
             params.append(self.colscale_opts)
         if self.letter_data is not None:
-            letter_layer = geom_polygon(data=self.letter_data, mapping=aes(x='x', y='y', group='group_by'),
-                                        fill=self.__font_col, **self.__kwargs)
+            if self.__scheme_applied == 'BACKGROUND':
+                letter_layer = geom_polygon(data=self.letter_data, mapping=aes(x='x', y='y', group='group_by'),
+                                            fill=self.__no_scheme_col, **self.__kwargs)
+            else:
+                letter_layer = geom_polygon(data=self.letter_data,
+                                            mapping=aes(x='x', y='y', group='group_by', fill='group'),
+                                            **self.__kwargs)
             params.append(letter_layer)
         params.extend([self.scale_x_continuous, self.scale_y_continuous, self.ylab, self.xlab])
         gg = gg + params

diff --git a/tests/test_align.py b/tests/test_align.py
@@ -10,7 +10,8 @@ def test_geom_alignedSeq():
     ggplot() + geom_alignedSeq(data=seqs, font=None) + theme_seq()
     ggplot() + geom_alignedSeq(data=seqs_dna['MA0013.1'], font=None) + theme_seq()
     ggplot() + geom_alignedSeq(data=seqs) + theme_seq() + coord_fixed()
+    ggplot() + geom_alignedSeq(data=seqs, scheme_applied='LETTER', no_scheme_col='white') + theme_seq() + coord_fixed()
     ggplot() + geom_alignedSeq(data=seqs_dna['MA0013.1']) + theme_seq()
     ggplot() + geom_alignedSeq(data=seqs,
                                seq_names=['test-aaa', 'test-b', 'test-c', 'test-d', 'test-e', 'test-ff']) + theme_seq()
-    ggplot() + geom_alignedSeq(data=seqs, font_col='black', bg_col_scheme=None) + theme_seq()
+    ggplot() + geom_alignedSeq(data=seqs, no_scheme_col='black', col_scheme=None) + theme_seq()
diff --git a/tests/test_index.py b/tests/test_index.py
@@ -17,11 +17,15 @@ def test_plot_an_aligned_sequences():
 
 
 def test_no_background_color():
-    ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], font_col='black', bg_col_scheme=None) + theme_seq() + coord_fixed()
+    ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], no_scheme_col='black', col_scheme=None) + theme_seq() + coord_fixed()
 
 
 def test_no_sequence_letter():
-    ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], font=None) + theme_seq() + coord_fixed()
+    ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], font=None, border_col=None) + theme_seq() + coord_fixed()
+
+
+def test_switch_color_schme():
+    ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], no_scheme_col='white', scheme_applied='LETTER') + theme_seq() + coord_fixed()
 
 
 def test_tagging_sequences():

diff --git a/tests/test_paper.py b/tests/test_paper.py
@@ -70,7 +70,7 @@
 O.save('Fig. 2O.png')
 P = ggplot() + geom_seqBar(seqs_dna['MA0013.1'], font=None) + ggtitle('P') + theme_seq()# Fig. 2P
 P.save('Fig. 2P.png')
-Q = ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], font_col='black', bg_col_scheme=None) + ggtitle(
+Q = ggplot() + geom_alignedSeq(seqs_dna['MA0013.1'], no_scheme_col='black', col_scheme=None) + ggtitle(
     'Q') + theme_seq() + coord_fixed()  # Fig. 2G
 Q.save('Fig. 2Q.png')
 from plotnine import scale_y_continuous