Skip to content

Commit

Permalink
clean_remove cancer_typoe refitting
Browse files Browse the repository at this point in the history
  • Loading branch information
@cdbope
cdbope committed May 27, 2024
1 parent 20dd6ac commit 2384b91
Showing 1 changed file with 2 additions and 133 deletions.
135 changes: 2 additions & 133 deletions mutational_starsign/cli.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -278,15 +278,14 @@ def refit(matrix_file: Annotated[str, typer.Argument(help='Tab separated matrix
# genotyped: Annotated[bool, typer.Argument(help="True if the VCF file has genotype information for many samples")] = False, output_folder: str = 'output/',
# cancer_type: Annotated[str,typer.Argument(help="Cancer type abbreviation, eg.: bcla, brca, chol, gbm, lgg, cesc, coad, esca, uvm, hnsc, kich, kirp, kirc, lihc, luad, lusc, dlbc, laml, ov, paad, prad, sarc, skcm, stad, thca, ucec")] = None):
'''
Mutational SIgnatures Refit Parameters \n
Mutational Signatures Refit Parameters \n
n_bootstraps: \n
matrix_file: str \n
signature_file: str \n
output_file_exposure: str \n
opportunity_file: str \n
numeric_chromosomes,bool: True if chromosome names in vcf are '1', '2', '3'. False if 'chr1', 'chr2', 'chr3' \n
genotyped,bool:True if the VCF file has genotype information for many samples \n
cancer_type, str: bcla, brca, chol, gbm, lgg, cesc, coad, esca, uvm, hnsc, kich, kirp, kirc, lihc, luad, lusc, dlbc, laml, ov, paad, prad, sarc, skcm, stad, thca, ucec
'''
sig_name = Path(signature_file).stem
Expand Down Expand Up @@ -331,7 +330,7 @@ def refit(matrix_file: Annotated[str, typer.Argument(help='Tab separated matrix
S = S[desired_order]
S = S.to_numpy().astype(float)
M = M.to_numpy().astype(float)
S, index_signature = select_signature_matrix(S, cancer_type, index_signature)
# S, index_signature = select_signature_matrix(S, cancer_type, index_signature)
O = read_opportunity(M, opportunity_file)
lambd = 0.7

Expand Down Expand Up @@ -373,134 +372,10 @@ def refit(matrix_file: Annotated[str, typer.Argument(help='Tab separated matrix
plot_top_five.savefig(f"{output_folder}/{run_name}.png", dpi=600)
plot_variance = cohort_violin(E)
plot_variance.savefig(f"{output_folder}/{run_name}.png", dpi=600)
# sum_expo.to_csv(f'{output_folder}/average_exposure_cohort.txt', index=index_signature,
# header=True,
# sep='\t')
np.savetxt(f'{output_folder}/average_{run_name}.txt', np.array(sum_expo_t))
print("--- %s seconds ---" % (time.time() - start_time))


def select_signature_matrix(S, cancer_type, index_signature):
if cancer_type is not None:
true_order = 'Type SBS1 SBS2 SBS3 SBS4 SBS5 SBS6 SBS7a SBS7b SBS7c SBS7d SBS8 SBS9 SBS10a SBS10b SBS10c SBS10d SBS11 SBS12 SBS13 SBS14 SBS15 SBS16 SBS17a SBS17b SBS18 SBS19 SBS20 SBS21 SBS22a SBS22b SBS23 SBS24 SBS25 SBS26 SBS27 SBS28 SBS29 SBS30 SBS31 SBS32 SBS33 SBS34 SBS35 SBS36 SBS37 SBS38 SBS39 SBS40a SBS40b SBS40c SBS41 SBS42 SBS43 SBS44 SBS45 SBS46 SBS47 SBS48 SBS49 SBS50 SBS51 SBS52 SBS53 SBS54 SBS55 SBS56 SBS57 SBS58 SBS59 SBS60 SBS84 SBS85 SBS86 SBS87 SBS88 SBS89 SBS90 SBS91 SBS92 SBS93 SBS94 SBS95 SBS96 SBS97 SBS98 SBS99'.split()[
1:]
# assert index_signature.tolist() != true_order, (f'The order of the signatures in the signature file is not the same as cosmic 3.4. Cannot do automatic selection of {cancer_type} signatures', index_signature.tolist(), true_order)
assert index_signature == true_order, (
f'The order of the signatures in the signature file is not the same as cosmic 3.4. You can download the file at https://cancer.sanger.ac.uk/cosmic/download/cosmic. Cannot do automatic selection of {cancer_type} signatures',
index_signature, true_order)
if cancer_type == 'bcla':
index = [0, 1, 3, 4, 18]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'brca':
index = [0, 1, 2, 4, 18, 24, 37, 41, 50]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'chol':
index = [0, 1, 4, 18, 24, 47, 48, 49, 67]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'gbm':
index = [0, 4, 37, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'lgg':
index = [0, 4]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'cesc':
index = [0, 1, 4, 18]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'coad':
index = [0, 4, 20, 47, 48, 49, 53]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'esca':
index = [0, 2, 4, 18, 22, 23, 24, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'uvm':
index = [0, 4, 47, 48, 49, 60, 61]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'hnsc':
index = [0, 1, 3, 4, 18, 47, 48, 49, 54]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'kich':
index = [0, 1, 18, 36, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'kirp':
index = [0, 1, 4, 18, 54, 58]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'kirc':
index = [0, 4, 28, 29, 47, 48, 49, 50]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'lihc':
index = [0, 2, 3, 4, 17, 21, 24, 28, 29, 36, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'luad':
index = [0, 1, 3, 4, 18, 47, 48, 54]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'lusc':
index = [0, 1, 3, 4, 18, 54]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'dlbc':
index = [1, 2, 4, 5, 11, 18, 22, 23, 41, 43, 44, 47, 48, 49, 65]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'dlbc':
index = [0, 1, 2, 4, 5, 11, 18, 22, 23, 41, 43, 44, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'laml':
index = [0, 4, 52]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'ov':
index = [0, 1, 2, 4, 18, 24, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'paad':
index = [0, 1, 2, 4, 18, 22, 23, 24, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'prad':
index = [0, 2, 4, 24, 47, 48, 49, 67]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'sarc':
index = [0, 4, 6, 7, 8, 9, 24, 68]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'skcm':
index = [0, 4, 6, 7, 8, 9, 45, 47, 48, 49, 54, 58]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'stad':
index = [0, 1, 2, 18, 20, 22, 23, 24, 26, 47, 48, 49]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'thca':
index = [0, 1, 4, 18, 47, 48, 49, 52, 54, 58, 62]
S = S[index]
index_signature = [index_signature[i] for i in index]
elif cancer_type == 'ucec':
index = [0, 1, 4, 12, 13, 14, 15, 18, 19, 20, 35, 53]
S = S[index]
index_signature = [index_signature[i] for i in index]
else:
raise ValueError(
f'Unknown cancer type {cancer_type}. Valid cancer types are: bcla, brca, chol, gbm, lgg, cesc, coad, esca, uvm, hnsc, kich, kirp, kirc, lihc, luad, lusc, dlbc, laml, ov, paad, prad, sarc, skcm, stad, thca, ucec')
return S, index_signature


def read_opportunity(M, opportunity_file):
n_samples = len(M)
Expand Down Expand Up @@ -567,12 +442,6 @@ def read_opportunity(M, opportunity_file):
O = np.ones((n_samples, n_mutations), dtype=float)
O = O / np.amin(O).sum(axis=-1, keepdims=True)

# The exposure is normalize to have the same proportion to the catalogue matrix
# normalized_vector1 = O / np.linalg.norm(O)
# min_value_vector2 = np.min(M)
# max_value_vector2 = np.max(M)
# O = normalized_vector1 * (max_value_vector2 - min_value_vector2) + min_value_vector2

assert O.shape == (n_samples, n_mutations), f'{O.shape} != {(n_samples, n_mutations)}'
return O

Expand Down

0 comments on commit 2384b91

Please sign in to comment.