def evaluate(chosenAA):
#get all possible triplet codons for the desired AA
possible_triplets = []
for aa in chosenAA:
possible_triplets.append(dna.GetCodons(aa, separate=True))
first, second, third = sumupcodons(possible_triplets) #takes the triplets and splits them up into their first, second and third positions
print('allcodons', (first, second, third))
degenerate1 = degenerate(first) #Gets degenerate codon that represents all bases at position 1
print('degenerate1', degenerate1)
degenerate2 = degenerate(second) #Gets degenerate codon that represents all bases at position 2
print('degenerate2', degenerate2)
degenerate3 = degenerate(third) #Gets degenerate codon that represents all bases at position 3
print('degenerate3', degenerate3)
if degenerate1 == 'N' and degenerate2 == 'N' and degenerate3 == 'N':
triplet = degenerate1 + degenerate2 + 'K' #for this special case I want NNK, not NNN
else:
triplet = degenerate1 + degenerate2 + degenerate3 # summing up
##triplet holds the degenerate triplet codon
###now I just need to convert this to a list of real codons and then check to which aa they match
Realcodons = dna.combine([dna.UnAmb(triplet[0]), dna.UnAmb(triplet[1]), dna.UnAmb(triplet[2])]) #condense the different codons for position 1, 2, and 3 to a list of triplets
ResultingAA = []
for codon in Realcodons:
ResultingAA.append(dna.Translate(codon)) #Check which AA these codons code for
TargetAA, OffTargetAA = chosenvsresulting(chosenAA, ResultingAA) #Check which of these AA were desired and which were not
return (triplet, TargetAA, OffTargetAA)
def find_degenerate(self, AA_list):
'''
Method for finding an degenerate codon encoding a list of desired amino acids.
The method finds the codon(s) with fewest off-target amino acids.
To reduce redundancy, the method then goes through all the best codons
(they all have the same number of off-target amino acids) and finds the one with the lowest number of codons.
If there are still more than one which are equivalent, the method then picks one WITHOUT a stop codon.
The input is a list of upper case amino acids in single-letter code.
The valid values are: FLSYCWPHERIMTNKVADQG*
The output is a tuple of the best degenerate codon and the off-target amino acids.
The degenerate codon is a string of three of the following characters: GATCRYWSMKHBVDN
The off-target amino acids is a list of upper case amino acids in single letter code.
'''
#make sure input is OK
assert all([
s in 'FLSYCWPHERIMTNKVADQG*U' for s in AA_list
]), 'Error, one or more of the amino acids %s are not valid.' % AA_list
#get all codons for chosen amino acids
regular_triplets = [
dna.GetCodons(aa,
table=self.getTable(),
separate=True,
exclude=True) for aa in AA_list
]
#some of the codons are list of lists (happens when the amino acid has codons at different parts of the codon circle)
#I need to flatten this into separate lists with which go on further
regular_triplets = self.flatten_codon_list(regular_triplets)
best_score = None
all_alternatives = [] #to save the result of all possible triplets
for codon_list in regular_triplets:
#get all nucleotides for first, second and third position while retaining list structure
first, second, third = self.sumupcodons(codon_list)
#check which degenerate nucleotide can be used to find at least one match in each of the lists
possible_triplets = dna.combine([
dna.commonNuc(first),
dna.commonNuc(second),
dna.commonNuc(third)
])
#now go through them and see which is best
for triplet in possible_triplets:
#convert the triplet back to a list of real codons
Realcodons = dna.combine(
[
dna.UnAmb(triplet[0]),
dna.UnAmb(triplet[1]),
dna.UnAmb(triplet[2])
]
) #condense the different codons for position 1, 2, and 3 to a list of triplets
#Check which AA these codons code for
ResultingAA = [
dna.Translate(codon, table=self.getTable())
for codon in Realcodons
]
#compare which amino acids were desired with the ones resulting from the degenerate codon
offtarget = sorted(
self.extra_list_elements(AA_list, ResultingAA))
#add to all options
if any([True for s in all_alternatives if s[0] == triplet
]) is False:
all_alternatives.append([triplet] + AA_list + offtarget)
#if there are fewer off-target amino acids with the new codon, keep it
if len(offtarget) < best_score or best_score == None:
best_score = len(offtarget)
good_triplets = []
good_triplets.append(triplet)
elif len(offtarget) == best_score:
good_triplets.append(triplet)
#the saved triplets all have the same number of off-target amino acids, now keep the one with the lowest number of codons (to reduce ambiguity)
best_triplet = None #for storing best degenerate triplet
best_offtarget = None #for storing the off-target AA of the best triplet
best_score = None #for storing the length of the off-target list
alternatives = [
] #for saving alternative triplets and their encoded amino acids
for triplet in good_triplets:
#convert the triplet back to a list of real codons
Realcodons = dna.combine(
[
dna.UnAmb(triplet[0]),
dna.UnAmb(triplet[1]),
dna.UnAmb(triplet[2])
]
) #condense the different codons for position 1, 2, and 3 to a list of triplets
#Check which AA these codons code for
ResultingAA = [
dna.Translate(codon, table=self.getTable())
for codon in Realcodons
]
#compare which amino acids were desired with the ones resulting from the degenerate codon
offtarget = sorted(self.extra_list_elements(AA_list, ResultingAA))
#save alternatives stats
if any([True for s in alternatives if s[0] == triplet]) is False:
alternatives.append([triplet] + AA_list + offtarget)
#save the stats in case there are fewer codons
if len(Realcodons) < best_score or best_score == None:
#save stats
best_score = len(Realcodons)
best_triplet = triplet
best_offtarget = offtarget
#if another codon has same stats as the previous best one, replace the previous codon if it has an off-target stop
elif len(Realcodons) == best_score and '*' in best_offtarget:
#save stats
best_score = len(Realcodons)
best_triplet = triplet
best_offtarget = offtarget
return best_triplet, best_offtarget, alternatives, all_alternatives
def find_degenerate(self, AA_list): ''' Method for finding an ambiguous codon encoding a list of desired amino acids. The method finds the codon with fewest off-target amino acids. The input is a list of upper case amino acids in the single-letter code. The valid values are: FLSYCWPHERIMTNKVADQG* The output is a tuple of the best ambiguous codon and the off-target amino acids. The ambiguous codon is a string of three of the following characters: GATCRYWSMKHBVDN The off-target amino acids is a list of upper case amino acids in single letter code. ''' #make sure input is OK assert all([s in 'FLSYCWPHERIMTNKVADQG*' for s in AA_list]), 'Error, one or more of the amino acids %s are not valid.' % AA_list #get all codons for chosen amino acids regular_triplets = [dna.GetCodons(aa, table=self.getTable(), separate=True) for aa in AA_list] #some of the codons are list of lists (happens when the amino acid has codons at different parts of the codon circle) #I need to flatten this into separate lists with which go on further regular_triplets = self.flatten_codon_list(regular_triplets) best_score = None for codon_list in regular_triplets: #get all nucleotides for first, second and third position while retaining list structure first, second, third = self.sumupcodons(codon_list) #check which degenerate nucleotide can be used to find at least one match in each of the lists possible_triplets = dna.combine([dna.commonNuc(first), dna.commonNuc(second), dna.commonNuc(third)]) #now go through them and see which is best for triplet in possible_triplets: #convert the triplet back to a list of real codons Realcodons = dna.combine([dna.UnAmb(triplet[0]), dna.UnAmb(triplet[1]), dna.UnAmb(triplet[2])]) #condense the different codons for position 1, 2, and 3 to a list of triplets #Check which AA these codons code for ResultingAA = [dna.Translate(codon, table=self.getTable()) for codon in Realcodons] #compare which amino acids were desired with the ones resulting from the degenerate codon offtarget = sorted(self.extra_list_elements(AA_list, ResultingAA)) #if there are fewer off-target amino acids with the new codon, keep it if len(offtarget) < best_score or best_score == None: best_score = len(offtarget) good_triplets = [] good_triplets.append(triplet) elif len(offtarget) == best_score: good_triplets.append(triplet) #the saved triplets all have the same number of off-target amino acids, now keep the one with the lowest number of codons for each AA best_triplet = None #for storing best degenerate triplet best_offtarget = None #for storing the off-target AA of the best triplet best_score = None #for storing the length of the off-target list for triplet in good_triplets: #convert the triplet back to a list of real codons Realcodons = dna.combine([dna.UnAmb(triplet[0]), dna.UnAmb(triplet[1]), dna.UnAmb(triplet[2])]) #condense the different codons for position 1, 2, and 3 to a list of triplets #save the stats in case there are fewer codons if len(Realcodons) < best_score or best_score == None: #Check which AA these codons code for ResultingAA = [dna.Translate(codon, table=self.getTable()) for codon in Realcodons] #compare which amino acids were desired with the ones resulting from the degenerate codon offtarget = sorted(self.extra_list_elements(AA_list, ResultingAA)) #save stats best_score = len(Realcodons) best_triplet = triplet best_offtarget = offtarget return best_triplet, best_offtarget
