remove find matches functions

algorithms/free_shift_alignment.py

@@ -23,28 +23,7 @@
 GAP = -10
 
 
-def fill_matches(score_matrix, ls1, ls2, seq1, seq2):
-    """creates a substitution matrix.
-    Starts with second row and column.
-
-    Args:
-        score_matrix (np matrix): initialized matrix
-        ls1 (int): len first sequence to alignt
-        ls2 (int): len second sequence to alignt
-        seq1 (str): first sequence to alignt
-        seq2 (str): second sequence to alignt
-
-    Returns:
-        numpy matrix: substitution matrix inside
-    initialized matrix
-    """
-    for y in range(1, ls2):
-        for x in range(1, ls1):
-            score_matrix[y][x] = MATCH if seq1[x] == seq2[y] else MISMATCH
-    return score_matrix
-
-
-def recalculate_scorematrix(score_matrix, ls1, ls2):
+def calculate_scorematrix(score_matrix, ls1, ls2, seq1, seq2):
     """creates a scorematrix with help of the
     substitution matrix and rewards/ penalties
 
@@ -53,15 +32,18 @@ def recalculate_scorematrix(score_matrix, ls1, ls2):
         initialized matrix
         ls1 (int): len first sequence to alignt
         ls2 (int): len second sequence to alignt
+        seq1 (str): first sequence to alignt
+        seq2 (str): second sequence to alignt
 
     Returns:
         numpy matrix: scorematrix
     """
     # start at second row and column -> skips initialized cells
     for y in range(1, ls2):
         for x in range(1, ls1):
+            match_value = MATCH if seq1[x] == seq2[y] else MISMATCH
             score_matrix[y][x] = max(
-                score_matrix[y-1][x-1] + score_matrix[y][x],
+                score_matrix[y-1][x-1] + match_value,
                 score_matrix[y-1][x] + GAP,
                 score_matrix[y][x-1] + GAP
             )
@@ -239,8 +221,7 @@ def main(seq1='', seq2='', assembly=False):
     ls1 = len(seq1)
     ls2 = len(seq2)
     score_matrix = np.zeros((ls2, ls1), dtype="int16")
-    score_matrix = fill_matches(score_matrix, ls1, ls2, seq1, seq2)
-    score_matrix = recalculate_scorematrix(score_matrix, ls1, ls2)
+    score_matrix = calculate_scorematrix(score_matrix, ls1, ls2, seq1, seq2)
     max_border_coords = get_max_from_border(score_matrix)
     seq1_new, seq2_new = traceback(
         score_matrix, seq1, seq2, max_border_coords)

algorithms/multiple_sequence_alignment.py

@@ -25,34 +25,6 @@
 GAP = -2
 
 
-def fill_matches(score_matrix, ls1, ls2, ls3, seq1, seq2, seq3):
-    """creates a substitution matrix inside
-    initialized matrix.
-    Starts with second row and column.
-
-    Args:
-        score_matrix (np matrix): initialized matrix
-        ls1 (int): len first sequence to alignt
-        ls2 (int): len second sequence to alignt
-        ls3 (int): len third sequence to alignt
-        seq1 (str): first sequence to alignt
-        seq2 (str): second sequence to alignt
-        seq2 (str): third sequence to alignt
-
-    Returns:
-        numpy matrix: substitution matrix inside
-    initialized matrix
-    """
-    for z in range(ls3):
-        for y in range(ls2):
-            for x in range(ls1):
-                if ((seq1[x] == seq2[y]) or (seq1[x] == seq3[z]) or (seq2[y] == seq3[z])):
-                    score_matrix[z][y][x] = MATCH
-                else:
-                    score_matrix[z][y][x] = MISMATCH
-    return score_matrix
-
-
 def initialize_matrix(score_matrix, ls1, ls2, ls3):
     """fillst first row and column with
     given penalties
@@ -76,7 +48,7 @@ def initialize_matrix(score_matrix, ls1, ls2, ls3):
     return score_matrix
 
 
-def recalculate_scorematrix(score_matrix, ls1, ls2, ls3):
+def calculate_scorematrix(score_matrix, ls1, ls2, ls3, seq1, seq2, seq3):
     """creates a scorematrix with help of the
     substitution matrix and rewards/ penalties
 
@@ -95,6 +67,11 @@ def recalculate_scorematrix(score_matrix, ls1, ls2, ls3):
         for y in range(ls2):
             for x in range(ls1):
                 coords = (x,y,z)
+                if ((seq1[x] == seq2[y]) or (seq1[x] == seq3[z]) or (seq2[y] == seq3[z])):
+                    match_value = MATCH
+                else:
+                    match_value = MISMATCH
+
                 # surfaces
                 if coords.count(0) == 1:
                     # front surface
@@ -301,17 +278,17 @@ def calc_similarity(alignment_output):
 
 def main():
     # change sequences below for your needs!
-    seq1 = ",AA,"
-    seq2 = ",AT,"
+    seq1 = ",AT,"
+    seq2 = ",AA,"
     seq3 = ",AA,"
 
     ls1 = len(seq1)
     ls2 = len(seq2)
     ls3 = len(seq3)
     score_matrix = np.zeros((ls3, ls2, ls1), dtype="int16")
-    score_matrix = fill_matches(score_matrix, ls1, ls2, ls3, seq1, seq2, seq3)
     score_matrix = initialize_matrix(score_matrix, ls1, ls2, ls3)
-    score_matrix = recalculate_scorematrix(score_matrix, ls1, ls2, ls3)
+    score_matrix = calculate_scorematrix(
+        score_matrix, ls1, ls2, ls3, seq1, seq2, seq3)
     seq1_new, seq2_new, seq3_new = traceback(score_matrix, ls1, ls2, ls3, seq1, seq2, seq3)
     alignment1_2, alignment1_3, alignment2_3 = output(
         seq1_new, seq2_new, seq3_new)

algorithms/needleman_wunsch.py

@@ -44,29 +44,7 @@ def initialize_matrix(score_matrix, ls1, ls2):
     return score_matrix
 
 
-def fill_matches(score_matrix, ls1, ls2, seq1, seq2):
-    """creates a substitution matrix inside
-    initialized matrix.
-    Starts with second row and column.
-
-    Args:
-        score_matrix (np matrix): initialized matrix
-        ls1 (int): len first sequence to alignt
-        ls2 (int): len second sequence to alignt
-        seq1 (str): first sequence to alignt
-        seq2 (str): second sequence to alignt
-
-    Returns:
-        numpy matrix: substitution matrix inside
-    initialized matrix
-    """
-    for y in range(1, ls2):
-        for x in range(1, ls1):
-            score_matrix[y][x] = MATCH if seq1[x] == seq2[y] else MISMATCH
-    return score_matrix
-
-
-def recalculate_scorematrix(score_matrix, ls1, ls2):
+def calculate_scorematrix(score_matrix, ls1, ls2, seq1, seq2):
     """creates a scorematrix with help of the
     substitution matrix and rewards/ penalties
 
@@ -75,15 +53,18 @@ def recalculate_scorematrix(score_matrix, ls1, ls2):
         initialized matrix
         ls1 (int): len first sequence to alignt
         ls2 (int): len second sequence to alignt
+        seq1 (str): first sequence to alignt
+        seq2 (str): second sequence to alignt
 
     Returns:
         numpy matrix: scorematrix
     """
     # start at second row and column -> skips initialized cells
     for y in range(1, ls2):
         for x in range(1, ls1):
+            match_value = MATCH if seq1[x] == seq2[y] else MISMATCH
             score_matrix[y][x] = max(
-                score_matrix[y-1][x-1] + score_matrix[y][x],
+                score_matrix[y-1][x-1] + match_value,
                 score_matrix[y-1][x] + GAP,
                 score_matrix[y][x-1] + GAP
                 )
@@ -195,8 +176,7 @@ def main(seq1='', seq2=''):
     ls2 = len(seq2)
     score_matrix = np.zeros((ls2, ls1), dtype="int16")
     score_matrix = initialize_matrix(score_matrix, ls1, ls2)
-    score_matrix = fill_matches(score_matrix, ls1, ls2, seq1, seq2)
-    score_matrix = recalculate_scorematrix(score_matrix, ls1, ls2)
+    score_matrix = calculate_scorematrix(score_matrix, ls1, ls2, seq1, seq2)
     seq1_new, seq2_new = traceback(score_matrix, ls1, ls2, seq1, seq2)
     alignment_output = output(seq1_new, seq2_new)
     sim_tup = calc_similarity(alignment_output)

algorithms/smith_waterman.py

@@ -24,29 +24,7 @@
 GAP = -2
 
 
-def fill_matches(score_matrix, ls1, ls2, seq1, seq2):
-    """creates a substitution matrix inside
-    initialized matrix.
-    Starts with second row and column.
-
-    Args:
-        score_matrix (np matrix): initialized matrix
-        ls1 (int): len first sequence to alignt
-        ls2 (int): len second sequence to alignt
-        seq1 (str): first sequence to alignt
-        seq2 (str): second sequence to alignt
-
-    Returns:
-        numpy matrix: substitution matrix inside
-    initialized matrix
-    """
-    for y in range(1, ls2):
-        for x in range(1, ls1):
-            score_matrix[y][x] = MATCH if seq1[x] == seq2[y] else MISMATCH
-    return score_matrix
-
-
-def recalculate_scorematrix(score_matrix, ls1, ls2):
+def calculate_scorematrix(score_matrix, ls1, ls2, seq1, seq2):
     """creates a scorematrix with help of the
     substitution matrix and rewards/ penalties
 
@@ -55,18 +33,22 @@ def recalculate_scorematrix(score_matrix, ls1, ls2):
         initialized matrix
         ls1 (int): len first sequence to alignt
         ls2 (int): len second sequence to alignt
+        seq1 (str): first sequence to alignt
+        seq2 (str): second sequence to alignt
 
     Returns:
         numpy matrix: scorematrix
     """
+    # start at second row and column -> skips initialized cells
     for y in range(1, ls2):
         for x in range(1, ls1):
+            match_value = MATCH if seq1[x] == seq2[y] else MISMATCH
             score_matrix[y][x] = max(
                 0,
-                score_matrix[y-1][x-1] + score_matrix[y][x],
+                score_matrix[y-1][x-1] + match_value,
                 score_matrix[y-1][x] + GAP,
                 score_matrix[y][x-1] + GAP
-                )
+            )
     return score_matrix
 
 
@@ -180,8 +162,7 @@ def main(seq1='', seq2=''):
     ls1 = len(seq1)
     ls2 = len(seq2)
     score_matrix = np.zeros((ls2, ls1), dtype="int16")
-    score_matrix = fill_matches(score_matrix, ls1, ls2, seq1, seq2)
-    score_matrix = recalculate_scorematrix(score_matrix, ls1, ls2)
+    score_matrix = calculate_scorematrix(score_matrix, ls1, ls2, seq1, seq2)
     max_coords = find_max_coordinates(score_matrix)
     seq1_new, seq2_new = traceback(score_matrix, seq1, seq2, max_coords)
     alignment_output = output(seq1_new, seq2_new)