fix

diann_lib_processing.py

@@ -5,7 +5,7 @@ import pyarrow.parquet as pq
 import pyarrow as pa
 import torch
 import matplotlib.pyplot as plt
-# from loess.loess_1d import loess_1d
+from loess.loess_1d import loess_1d
 
 from model.model import ModelTransformer
 from config import load_args
@@ -96,98 +96,94 @@ def predict(data_pred, model, output_path):
 
 
 if __name__ =='__main__':
-    # df = load_lib('spectral_lib/1-240711_ident_resistance_idbioriv_fluoroquinolones_conta_human_sang.parquet')
-    # df = extract_sequence(df)
-    # df.to_csv('spectral_lib/1-240711_ident_resistance_idbioriv_fluoroquinolones_conta_human_sang.csv')
-    # plt.hist(df['RT'])
-    # plt.savefig('test.png')
-    #
-    # df_2 = pd.read_csv('data_prosit/data.csv')
-    #
-    # plt.clf()
-    # plt.hist(df_2['irt'])
-    # plt.savefig('test2.png')
-    #
-    # df_2 = extract_sequence(df).reset_index(drop=True)
-    #
-    # pred = pd.read_csv('../output/out_lib_CITBASE_try_contaminant.csv')
-    #
-    # pred['seq']=pred['seq'].map(numerical_to_alphabetical_str)
-    #
-    # pred['Modified.Sequence']=pred['seq']
-    #
-    # result = pd.merge(df,pred[['Modified.Sequence','rt pred']],on='Modified.Sequence',how='left')
-    #
-    #
-    #
-    # #alignement
-    #
-    # ref = pd.read_csv('data_prosit/data_noc.csv')
-    # df_ISA = pd.read_csv('data_ISA/data_aligned_isa_noc.csv')
-    #
-    # dataset, reference, column_dataset, column_ref, seq_data, seq_ref = df_ISA, ref,  'irt_scaled', 'irt', 'sequence','sequence',
-    #
-    # dataset_ref=dataset[dataset['state']=='train']
-    # dataset_unique = dataset_ref[[seq_data,column_dataset]].groupby(seq_data).mean()
-    # print('unique',len(dataset_unique))
-    # reference_unique = reference[[seq_ref,column_ref]].groupby(seq_ref).mean()
-    # seq_ref = reference_unique.index
-    # seq_common = dataset_unique.index
-    # seq_ref = seq_ref.tolist()
-    # seq_common = seq_common.tolist()
-    #
-    # seq_ref = [tuple(l) for l in seq_ref]
-    # seq_common = [tuple(l) for l in seq_common]
-    #
-    # ind_dict_ref = dict((k, i) for i, k in enumerate(seq_ref))
-    # inter = set(ind_dict_ref).intersection(seq_common)
-    # print(len(inter))
-    #
-    # ind_dict_ref = [ind_dict_ref[x] for x in inter]
-    #
-    # indices_common = dict((k, i) for i, k in enumerate(seq_common))
-    # indices_common = [indices_common[x] for x in inter]
-    #
-    #
-    # rt_ref = reference_unique[column_ref][ind_dict_ref].reset_index()
-    # rt_data = dataset_unique[column_dataset][indices_common].reset_index()
-    #
-    # plt.scatter(rt_data[column_dataset].tolist(),rt_ref[column_ref].tolist(),s=0.1)
-    # plt.savefig('test.png')
-    #
-    # #présence de NAN qui casse le réalignement (solution temporaire : remplacer par 0.
-    # result['rt pred']=result['rt pred'].fillna(value=0)
-    # xout, yout, wout = loess_1d(np.array(rt_data[column_dataset].tolist()), np.array(rt_ref[column_ref].tolist()),
-    #                             xnew=result['rt pred'],
-    #                             degree=1,
-    #                             npoints=None, rotate=False, sigy=None)
-    #
-    #
-    # #writing results
-    #
-    # result['RT'] = yout
-    #
-    # result = result.drop('rt pred', axis=1)
-    #
-    # table = pa.Table.from_pandas(result)
-    #
-    # pq.write_table(table, 'spectral_lib/first_lib_contaminant_prosit_aligned.parquet')
-    #
+    df = load_lib('spectral_lib/1-240711_ident_resistance_idbioriv_fluoroquinolones_conta_human_sang.parquet')
 
+    df_2 = pd.read_csv('data_prosit/data.csv')
 
-    args = load_args()
+    plt.clf()
+    plt.hist(df_2['irt'])
+    plt.savefig('test2.png')
 
-    model = ModelTransformer(encoder_ff=args.encoder_ff, decoder_rt_ff=args.decoder_rt_ff,
-                             n_head=args.n_head, encoder_num_layer=args.encoder_num_layer,
-                             decoder_rt_num_layer=args.decoder_rt_num_layer, drop_rate=args.drop_rate,
-                             embedding_dim=args.embedding_dim, acti=args.activation, norm=args.norm_first, seq_length=30)
+    df_2 = extract_sequence(df).reset_index(drop=True)
 
-    if torch.cuda.is_available():
-        model = model.cuda()
+    pred = pd.read_csv('../output/out_transfer_prosit_isa_1-240711_ident_resistance_idbioriv_fluoroquinolones_conta_human_sang.csv')
 
-    model.load_state_dict(torch.load(args.model_weigh, weights_only=True))
+    pred['seq']=pred['seq'].map(numerical_to_alphabetical_str)
 
-    data_test = load_data(data_source=args.dataset_test, batch_size=args.batch_size, length=30, mode=args.split_test,
-                          seq_col=args.seq_test)
+    pred['Modified.Sequence']=pred['seq']
 
-    predict(data_test, model, args.output)
+    result = pd.merge(df,pred[['Modified.Sequence','rt pred']],on='Modified.Sequence',how='left')
+
+
+
+    #alignement
+
+    ref = pd.read_csv('data_prosit/data_noc.csv')
+    df_ISA = pd.read_csv('data_ISA/data_aligned_isa_noc.csv')
+
+    dataset, reference, column_dataset, column_ref, seq_data, seq_ref = df_ISA, ref,  'irt_scaled', 'irt', 'sequence','sequence',
+
+    dataset_ref=dataset[dataset['state']=='train']
+    dataset_unique = dataset_ref[[seq_data,column_dataset]].groupby(seq_data).mean()
+    print('unique',len(dataset_unique))
+    reference_unique = reference[[seq_ref,column_ref]].groupby(seq_ref).mean()
+    seq_ref = reference_unique.index
+    seq_common = dataset_unique.index
+    seq_ref = seq_ref.tolist()
+    seq_common = seq_common.tolist()
+
+    seq_ref = [tuple(l) for l in seq_ref]
+    seq_common = [tuple(l) for l in seq_common]
+
+    ind_dict_ref = dict((k, i) for i, k in enumerate(seq_ref))
+    inter = set(ind_dict_ref).intersection(seq_common)
+    print(len(inter))
+
+    ind_dict_ref = [ind_dict_ref[x] for x in inter]
+
+    indices_common = dict((k, i) for i, k in enumerate(seq_common))
+    indices_common = [indices_common[x] for x in inter]
+
+
+    rt_ref = reference_unique[column_ref][ind_dict_ref].reset_index()
+    rt_data = dataset_unique[column_dataset][indices_common].reset_index()
+
+    plt.scatter(rt_data[column_dataset].tolist(),rt_ref[column_ref].tolist(),s=0.1)
+    plt.savefig('test.png')
+
+    #présence de NAN qui casse le réalignement (solution temporaire : remplacer par 0.
+    result['rt pred']=result['rt pred'].fillna(value=0)
+    xout, yout, wout = loess_1d(np.array(rt_data[column_dataset].tolist()), np.array(rt_ref[column_ref].tolist()),
+                                xnew=result['rt pred'],
+                                degree=1,
+                                npoints=None, rotate=False, sigy=None)
+
+
+    #writing results
+
+    result['RT'] = yout
+
+    result = result.drop('rt pred', axis=1)
+
+    table = pa.Table.from_pandas(result)
+
+    pq.write_table(table, 'spectral_lib/1-240711_ident_resistance_idbioriv_fluoroquinolones_conta_human_sang_finetune_aligned.parquet')
+
+
+
+    # args = load_args()
+    #
+    # model = ModelTransformer(encoder_ff=args.encoder_ff, decoder_rt_ff=args.decoder_rt_ff,
+    #                          n_head=args.n_head, encoder_num_layer=args.encoder_num_layer,
+    #                          decoder_rt_num_layer=args.decoder_rt_num_layer, drop_rate=args.drop_rate,
+    #                          embedding_dim=args.embedding_dim, acti=args.activation, norm=args.norm_first, seq_length=30)
+    #
+    # if torch.cuda.is_available():
+    #     model = model.cuda()
+    #
+    # model.load_state_dict(torch.load(args.model_weigh, weights_only=True))
+    #
+    # data_test = load_data(data_source=args.dataset_test, batch_size=args.batch_size, length=30, mode=args.split_test,
+    #                       seq_col=args.seq_test)
+    #
+    # predict(data_test, model, args.output)

identification/result_extraction.py

@@ -41,7 +41,19 @@ def compare_error(path_1,path_2):
     plt.savefig('error2.png')
     return error_1,error_2
 
+def compare_with_db(path):
+    df = pd.read_csv(path, sep='\t', encoding='latin-1')
+    df_ref = pd.read_excel('250205_All_Peptides_panel_ID_+_RES.xlsx',names=['peptide','fonction'])
+    df2=df[df['Stripped.Sequence'].isin(df_ref['peptide'].to_list())]
+    corespondance = pd.merge(df2,df_ref,left_on='Stripped.Sequence',right_on='peptide',how="left")
+
+    return corespondance
+
 
 if __name__ == '__main__':
-    # compare_id('CITCRO_ANA_3/report_custom.tsv', 'CITCRO_ANA_3/report_first_lib.tsv', 'CITCRO_ANA_3/report_finetune.tsv','CITCRO_ANA_3')
-    e1,e2 = compare_error('CITCRO_ANA_3/report_custom.tsv', 'CITCRO_ANA_3/report_first_lib.tsv')
+    # compare_id('CITAMA_ANA_5/julie_custom_nolib.tsv', 'CITAMA_ANA_5/julie_base_nolib.tsv', 'CITAMA_ANA_5/julie_finetune_nolib.tsv','CITAMA_ANA_5_julie_no_lib')
+    # e1,e2 = compare_error('CITAMA_ANA_5/report_custom.tsv', 'CITCRO_ANA_3/report_first_lib.tsv')
\ No newline at end of file
+
+    cor_base = compare_with_db('CITAMA_ANA_5/julie_base_nolib.tsv')
+    cor_custom = compare_with_db('CITAMA_ANA_5/julie_custom_nolib.tsv')
+    cor_finetune = compare_with_db('CITAMA_ANA_5/julie_finetune_nolib.tsv')
\ No newline at end of file