initial commit

app.py

@@ -0,0 +1,284 @@
+import gradio as gr
+import pandas as pd
+from pathlib import Path
+import ast
+
+'''
+Causal Gene Discovery Model
+/home/ema30/zaklab/rare_disease_dx/checkpoints/aligner/04_30_22:13:29:55_lr_1e-05_val_simulated_pats.disease_split_val_sim_pats_kg_8.9.21_kg_losstype_gene_multisimilarity/all_udn_patients_kg_8.9.21_kgsolved_manual_baylor_nobgm_distractor_genes_5_candidates_mapped_only_genes
+
+Patients-Like-Me Model
+/home/ema30/zaklab/rare_disease_dx/checkpoints/patient_NCA/04_26_22:17:38:30_lr_5e-05_val_simulated_pats.disease_split_val_sim_pats_kg_8.9.21_kg_losstype_patient_patient_NCA/mygene2_all_sim_all_udn_patients_kg_8.9.21_kgsolved_with_phenotypes
+
+Disease Characterization Model
+/home/ema30/zaklab/rare_disease_dx/checkpoints/patient_NCA/05_13_22:08:00:32_lr_1e-05_val_simulated_pats.disease_split_val_sim_pats_kg_8.9.21_kg_losstype_pd_NCA/mygene2_all_sim_all_udn_patients_kg_8.9.21_kgsolved_with_phenotypes
+'''
+
+
+
+gene_scores_df = pd.read_csv('gene_discovery_scores.csv')
+exomiser_gene_scores_df = pd.read_csv('exomiser_gene_discovery_scores.csv')
+patient_scores_df = pd.read_csv('patients_like_me_scores.csv')
+dx_scores_df = pd.read_csv('dx_characterization_scores.csv')
+plm_attn_df = pd.read_csv('patients_like_me_scores_attn.csv')
+dx_attn_df = pd.read_csv('dx_characterization_scores_attn.csv')
+gene_attn_df = pd.read_csv('gene_discovery_scores_attn.csv')
+exomiser_gene_attn_df = pd.read_csv('exomiser_gene_discovery_scores_attn.csv')
+
+diseases_map = {'UDN-P1': 'POLR3-releated leukodystrophy', 'UDN-P2': 'Novel Syndrome', 'UDN-P3':'Coffin-Lowry syndrome' , 
+                'UDN-P4': 'automsomal recessive spastic paraplegia type 76', 'UDN-P5': 'atypical presentation of familial cold autoinflammatory syndrome', 
+                'UDN-P6': '*GATAD2B*-associated syndrome', 'UDN-P7': 'AR limb-girdle muscular atrophy type 2D', 'UDN-P8': '*ATP5PO*-related Leigh syndrome', 'UDN-P9': 'Spondyloepimetaphyseal dysplasia, Isidor-Toutain type'}
+genes_map = {'UDN-P3': 'RPS6KA3', 'UDN-P4': 'CAPN1', 'UDN-P5': 'NLRP12, RAPGEFL1', 'UDN-P6': 'GATAD2B', 'UDN-P7': 'SGCA', 'UDN-P8': 'ATP5P0', 'UDN-P9': 'RPL13'}
+
+
+
+def get_patient(patient_id, attn_df):
+    '''
+    Returns phenotypes, candidate genes, Causal gene, disease
+    '''
+    if patient_id in genes_map: gene = genes_map[patient_id]
+    else:
+        patient_gene_scores_df = gene_scores_df.loc[gene_scores_df['patient_id'] == patient_id]
+        gene = ', '.join(patient_gene_scores_df.loc[patient_gene_scores_df['correct_gene_label'] == 1, 'genes'].tolist())
+    
+    if patient_id in diseases_map: disease = diseases_map[patient_id]
+    else:
+        patient_dx_scores_df = dx_scores_df.loc[dx_scores_df['patient_id'] == patient_id] 
+        disease = ', '.join(patient_dx_scores_df.loc[patient_dx_scores_df['correct_label'] == 1, 'diseases'].tolist())
+    
+    patient_attn_df = attn_df.loc[attn_df['patient_id'] == patient_id]
+    phenotypes = ', '.join(patient_attn_df['phenotypes'].tolist())
+
+    patient_str = f'''
+    **Selected Patient:** {patient_id}<br>
+    **Causal Gene:** *{gene}*<br>
+    **Disease:** {disease}<br>
+    **Phenotypes:** {phenotypes}<br><br>
+    '''
+
+    return patient_str
+
+
+def read_file(filename):
+    with open(filename, 'r') as file:
+        f = file.read()
+    return f
+
+
+def causal_gene_discovery(patient_id, prioritization_type):
+    if prioritization_type == 'Variant Filtered':
+        scores_df = exomiser_gene_scores_df.loc[exomiser_gene_scores_df['patient_id'] == patient_id]
+    else:
+        scores_df = gene_scores_df.loc[gene_scores_df['patient_id'] == patient_id]
+
+
+
+    # read in gene scores
+    scores_df = scores_df.sort_values("similarities", ascending=False)
+    scores_df['similarities'] = scores_df['similarities'].round(3).astype(str)
+
+    # add links to gene cards
+    scores_df['genes'] = scores_df['genes'].apply(lambda x: f'<u>[{x}](https://www.genecards.org/cgi-bin/carddisp.pl?gene={x})</u>')
+
+    # bold/color causal gene
+    scores_df.loc[scores_df['correct_gene_label'] == 1, 'similarities'] = scores_df.loc[scores_df['correct_gene_label'] == 1, 'similarities'].apply(lambda x: f'<span style="color:green">**{x}**</span>')
+    scores_df.loc[scores_df['correct_gene_label'] == 1, 'genes'] = scores_df.loc[scores_df['correct_gene_label'] == 1, 'genes'].apply(lambda x: f'<span style="color:green">**{x}**</span>')
+
+    #filter df
+    scores_df = scores_df.drop(columns=['patient_id', 'correct_gene_label']).rename(columns={ 'similarities': 'SHEPHERD Score', 'genes': 'Candidate Genes'}) #'correct_gene_label' : 'Is Causal Gene',
+
+    #############
+    # Attention
+
+    #read in phenotype attention
+    if prioritization_type == 'Variant Filtered':
+        attn_df = exomiser_gene_attn_df.loc[exomiser_gene_attn_df['patient_id'] == patient_id]
+    else:
+        attn_df = gene_attn_df.loc[gene_attn_df['patient_id'] == patient_id]
+    attn_df = attn_df.sort_values("attention", ascending=False)
+    attn_df['attention'] = attn_df['attention'].round(4)
+    attn_df = attn_df.drop(columns=['patient_id', 'degrees'])
+
+    #############
+    # KG neighborhood
+    #image_loc = f'images/{patient_id}.png'
+    html_file = f'https://michellemli.github.io/test_html/{patient_id}.html'
+    kg_html = f'''<iframe id="igraph" scrolling="no" style="border:none;  width: 100%; height: 600px" seamless="seamless" src="{html_file}"></iframe>'''
+
+
+    #patient_info
+    patient = get_patient(patient_id, gene_attn_df)
+
+    return patient, scores_df, attn_df, kg_html
+
+
+def patients_like_me(patient_id, k=10):
+
+
+    scores_df = patient_scores_df.loc[patient_scores_df['patient_id'] == patient_id]
+    scores_df = scores_df.sort_values("similarities", ascending=False)
+
+    #scores_df['phenotypes'] ='PHEN'
+
+    # add links to disease pages
+    scores_df['disease_ids'] = scores_df['disease_ids'].apply(lambda x: f'(https://www.orpha.net/consor/cgi-bin/OC_Exp.php?lng=en&Expert={x})</u>')
+    scores_df['diseases'] = scores_df['diseases'].apply(lambda x: f'<u>[{x}]')
+    scores_df['diseases'] = scores_df['diseases'] + scores_df['disease_ids']
+
+    scores_df['genes'] = scores_df['genes'].apply(lambda x: f'<u>[{x}](https://www.genecards.org/cgi-bin/carddisp.pl?gene={x})</u>')
+
+
+
+    # bold/color patients with same causal gene
+    scores_df.loc[scores_df['correct_label'] == 1, 'candidate_patients'] = scores_df.loc[scores_df['correct_label'] == 1, 'candidate_patients'].apply(lambda x: f'<span style="color:green">**{x}**</span>')
+    scores_df.loc[scores_df['correct_label'] == 1, 'genes'] = scores_df.loc[scores_df['correct_label'] == 1, 'genes'].apply(lambda x: f'<span style="color:green">**{x}**</span>')
+    scores_df.loc[scores_df['correct_label'] == 1, 'diseases'] = scores_df.loc[scores_df['correct_label'] == 1, 'diseases'].apply(lambda x: f'<span style="color:green">**{x}**</span>')
+
+    scores_df = scores_df.drop(columns=['patient_id', 'similarities', 'correct_label', 'disease_ids']).rename(columns={'candidate_patients': 'Candidate Patient', 'genes': 'Candidate Patient\'s Gene', 'diseases': 'Candidate Patient\'s Disease' }) #'phenotypes': 'Candidate Patient\'s Phenotypes'
+    scores_df = scores_df.head(k)
+
+
+    #read in phenotype attention
+    attn_df = plm_attn_df.loc[plm_attn_df['patient_id'] == patient_id]
+    attn_df = attn_df.sort_values("attention", ascending=False)
+    attn_df['attention'] = attn_df['attention'].round(4)
+    attn_df = attn_df.drop(columns=['patient_id', 'degrees'])
+
+    #patient_info
+    patient = get_patient(patient_id, plm_attn_df)
+
+
+    return patient, scores_df, attn_df
+
+
+def disease_characterization(patient_id, k=10):
+
+
+    #TODO: limit # of rows
+    scores_df = dx_scores_df.loc[dx_scores_df['patient_id'] == patient_id]
+    scores_df = scores_df.sort_values("similarities", ascending=False)
+    scores_df = scores_df.head(k)
+
+    scores_df.loc[ scores_df['disease_ids'].str.contains('Coxa vara'), 'disease_ids'] = '2812'
+    scores_df.loc[ scores_df['disease_ids'].str.contains('Multiple epiphyseal dysplasia'), 'disease_ids'] = '2654'
+
+
+
+    scores_df['disease_ids'] = scores_df['disease_ids'].apply(lambda x: ast.literal_eval(x))
+    scores_df['type_disease_ids'] = scores_df['disease_ids'].apply(lambda x: type(x))
+
+    scores_df.loc[scores_df['type_disease_ids'] == list, 'disease_ids'] = scores_df.loc[scores_df['type_disease_ids'] == list, 'disease_ids'].apply(lambda x: x[0])
+
+
+    # add links to disease pages
+    scores_df['disease_ids'] = scores_df['disease_ids'].apply(lambda x: f'(https://www.orpha.net/consor/cgi-bin/OC_Exp.php?lng=en&Expert={x})</u>')
+    scores_df['diseases'] = scores_df['diseases'].apply(lambda x: f'<u>[{x}]')
+    scores_df['diseases'] = scores_df['diseases'] + scores_df['disease_ids']
+
+    # one disease couldn't map to orphanet
+    scores_df.loc[ scores_df['disease_ids'].str.contains('33657'), 'diseases'] = '<u>[leukodystrophy, hypomyelinating, 20](https://www.omim.org/entry/619071)</u>'
+    scores_df.loc[ scores_df['disease_ids'].str.contains('2654'), 'diseases'] = '<u>[Multiple epiphyseal dysplasia](https://www.orpha.net/consor/cgi-bin/OC_Exp.php?lng=EN&Expert=251)</u>'
+    scores_df.loc[ scores_df['disease_ids'].str.contains('2812'), 'diseases'] = '<u>[Coxa vara](https://omim.org/entry/122750)</u>'
+
+    
+
+    scores_df = scores_df.drop(columns=['patient_id', 'similarities', 'correct_label', 'disease_ids','type_disease_ids']).rename(columns={'diseases' : 'Disease'})
+
+
+
+    #read in phenotype attention
+    attn_df = dx_attn_df.loc[dx_attn_df['patient_id'] == patient_id]
+    attn_df = attn_df.sort_values("attention", ascending=False)
+    attn_df['attention'] = attn_df['attention'].round(4)
+    attn_df = attn_df.drop(columns=['patient_id', 'degrees'])
+
+    #patient_info
+    patient = get_patient(patient_id, dx_attn_df)
+
+
+
+    return patient, scores_df, attn_df
+
+def get_umap(umap_type):
+        # get UMAP
+    if umap_type == 'disease':
+        html_file = 'https://michellemli.github.io/test_html/shepherd_disease_characterization_umap.html'
+        #html_file = read_file('images/udn_orphafit_patient_umap_nneigh=50_mindist=0.9_spread=1.0colored_by_disease_category.html')
+    elif umap_type == 'patient':
+        html_file = 'https://michellemli.github.io/test_html/shepherd_patient_umap.html'
+
+    else:
+        raise NotImplementedError
+
+
+    # return f"""<iframe style="width: 100%; height: 480px" name="result" allow="midi; 
+    # display-capture; encrypted-media;" sandbox="allow-modals allow-forms 
+    # allow-scripts allow-same-origin allow-popups 
+    # allow-top-navigation-by-user-activation allow-downloads" allowfullscreen="" 
+    # allowpaymentrequest="" frameborder="0" srcdoc='{html_file}'></iframe>"""
+    return f'''<embed style="border: none;" src="{html_file}" dpi="300" width="100%" height="750px" />'''
+
+    #return f'''<iframe id="igraph" scrolling="no" style="border:none;  width: 100%; height: 750px" seamless="seamless" src="{html_file}"></iframe>'''
+
+
+with gr.Blocks() as demo: #css="#gene_attn_accordion {text-align: center}" css="kg_neigh {width: 70%}"
+    gr.Markdown('<center><h1>AI-assisted Rare Disease Diagnosis with SHEPHERD</h1></center>')
+    #gr.Markdown('<center><h2>A few SHot Explainable Predictor for Hard-to-diagnosE Rare Diseases</h2></center>')
+
+    with gr.Tabs():
+        with gr.TabItem("Causal Gene Discovery"):
+            with gr.Column():
+                gr.Markdown('<center><h2>Select a patient to view SHEPHERD\'s predictions</h2></center>')
+                gene_dropdown = gr.Dropdown(choices=['UDN-P1', 'UDN-P2'],  label='Rare Disease Patients', type='value') #value='UDN-P1',
+                gene_radio = gr.Radio(choices=['Expert Curated', 'Variant Filtered'], value='Expert Curated', label='Type of Gene List')
+                patient_info = gr.Markdown() #get_patient('UDN-P1')
+
+                with gr.Accordion(label=f'SHEPHERD\'s Ranking of Patient\'s Candidate Genes', open=True, elem_id='gene_accordion'):
+                    #gr.Markdown(f'<center><h3>SHEPHERD\'s Ranking of Patient\'s Candidate Genes</h3></center>')
+                    gr.Markdown('The patient\'s causal gene (i.e. gene harboring a variant that explains the patient\'s symptoms) is colored in green.')
+                    gene_dataframe = gr.Dataframe(max_rows=5,  elem_id="gene_df", datatype = 'markdown', headers=['Candidate Genes', 'SHEPHERD Score' ], overflow_row_behaviour='paginate') # label='Candidate Genes', show_label=False,
+                with gr.Accordion(label=f'SHEPHERD\'s Attention to Patient\'s Phenotypes', open=False, elem_id='gene_attn_accordion'):
+                    #gr.Markdown(f'<center><h3>SHEPHERD\'s Attention to Patient\'s Phenotypes</h3></center>')
+                    gene_attn_dataframe = gr.Dataframe(max_rows=5,  elem_id="gene_attn_df", headers=['Phenotypes', 'Attention' ], overflow_row_behaviour='paginate') # label='Candidate Genes', show_label=False,
+                with gr.Accordion(label=f'Visualization of Patient\'s Neighborhood in the Knowledge Graph', open=False, elem_id='kg_neigh_accordion'):
+                    #kg_neighborhood_image = gr.Image(elem_id='kg_neigh')#.style(height=200, width=200)
+                    kg_neighborhood_image = gr.HTML(elem_id = 'kg_patient_neighborhood')
+
+                #gene_button = gr.Button("Go")
+
+        with gr.TabItem("Patients Like Me"):
+            gr.HTML(get_umap('patient'))
+            gr.Markdown('<center><h2>Select a patient to view SHEPHERD\'s predictions</h2></center>')
+            patient_dropdown = gr.Dropdown(choices=['UDN-P3','UDN-P4','UDN-P5','UDN-P6'],  label='Rare Disease Patients', type='value')
+            p_patient_info = gr.Markdown() 
+            with gr.Accordion(label=f'Most Similar Patients according to SHEPHERD', open=True, elem_id='pt_accordion'): #
+                patient_dataframe = gr.Dataframe(max_rows=10, datatype = 'markdown', show_label=False, elem_id="pat_df", headers=['Candidate Patient', 'Candidate Patient\'s Gene', 'Candidate Patient\'s Disease' ]) #'Candidate Patient\'s Phenotypes'
+                #patient_button = gr.Button("Go")
+            with gr.Accordion(label='SHEPHERD\'s Attention to Patient\'s Phenotypes', open=False, elem_id='pt_attn_accordion'):
+                pt_attn_dataframe = gr.Dataframe(max_rows=5,  elem_id="pt_attn_df", headers=['Phenotypes', 'Attention' ], overflow_row_behaviour='paginate')
+
+
+        with gr.TabItem("Disease Characterization"):
+            gr.HTML(get_umap('disease'))
+            gr.Markdown('<center><h2>Select a patient to view SHEPHERD\'s predictions</h2></center>')
+            dx_dropdown = gr.Dropdown(choices=['UDN-P7','UDN-P8','UDN-P9','UDN-P2'],  label='Rare Disease Patients', type='value')
+            dx_patient_info = gr.Markdown() 
+            with gr.Accordion(label='Top 10 Most Similar Diseases according to SHEPHERD', open=True, elem_id='pt_accordion'): #
+                dx_dataframe = gr.Dataframe(max_rows=10, datatype = 'markdown', show_label=False, elem_id="dx_df",  headers=['Diseases'])
+            with gr.Accordion(label='SHEPHERD\'s Attention to Patient\'s Phenotypes', open=False, elem_id='dx_attn_accordion'):
+                dx_attn_dataframe = gr.Dataframe(max_rows=5,  elem_id="dx_attn_df", headers=['Phenotypes', 'Attention' ], overflow_row_behaviour='paginate')
+
+            #dx_button = gr.Button("Go")
+
+    gene_dropdown.change(causal_gene_discovery, inputs=[gene_dropdown,gene_radio], outputs=[patient_info, gene_dataframe, gene_attn_dataframe, kg_neighborhood_image])
+    gene_radio.change(causal_gene_discovery, inputs=[gene_dropdown,gene_radio], outputs=[patient_info, gene_dataframe, gene_attn_dataframe, kg_neighborhood_image])
+
+    patient_dropdown.change(patients_like_me, inputs=patient_dropdown, outputs=[p_patient_info, patient_dataframe, pt_attn_dataframe])
+    dx_dropdown.change(disease_characterization, inputs=dx_dropdown, outputs=[dx_patient_info, dx_dataframe, dx_attn_dataframe])
+
+    #gene_dropdown.change(get_patient, inputs=gene_dropdown, outputs=patient_info)
+    #gene_button.click(causal_gene_discovery, inputs=gene_dropdown, outputs=[gene_dataframe,gene_attn_dataframe, kg_neighborhood_image])
+    #patient_button.click(patients_like_me, inputs=patient_dropdown, outputs=patient_dataframe)
+    #dx_button.click(disease_characterization, inputs=dx_dropdown, outputs=dx_dataframe)
+
+demo.launch( ) #server_port=50018, share=True

dx_characterization_scores.csv

@@ -0,0 +1,41 @@
+patient_id,diseases,similarities,correct_label,disease_ids
+UDN-P7,autosomal recessive limb-girdle muscular dystrophy type 2B,0.0004265006864443,0,[268]
+UDN-P7,GNE myopathy,0.0004263115406502,0,[602]
+UDN-P7,MYH7-related late-onset scapuloperoneal muscular dystrophy,0.0004259833076503,0,[437572]
+UDN-P7,"Emery-Dreifuss muscular dystrophy 2, autosomal dominant",0.0004257017571944,0,[264]
+UDN-P7,autosomal recessive limb-girdle muscular dystrophy type 2G,0.0004251024511177,0,[34514]
+UDN-P7,autosomal recessive limb-girdle muscular dystrophy type 2C,0.0004250344645697,0,[353]
+UDN-P7,myofibrillar myopathy 4,0.0004249221819918,0,[98912]
+UDN-P7,autosomal recessive limb-girdle muscular dystrophy type 2Q,0.0004244803567416,0,[254361]
+UDN-P7,distal myopathy with posterior leg and anterior hand involvement,0.0004238819237798,0,[63273]
+UDN-P7,autosomal recessive limb-girdle muscular dystrophy type 2I,0.000423846533522,0,[34515]
+UDN-P9,multiple epiphyseal dysplasia type 1,0.0004178489616606,0,[93308]
+UDN-P9,progressive pseudorheumatoid arthropathy of childhood,0.0004172090848442,0,[1159]
+UDN-P9,multiple epiphyseal dysplasia type 5,0.0004147714353166,0,[93311]
+UDN-P9,"metaphyseal chondrodysplasia, Spahr type",0.000414393376559,0,[2501]
+UDN-P9,Multiple epiphyseal dysplasia,0.0004142468678764,0,Multiple epiphyseal dysplasia
+UDN-P9,spondyloepiphyseal dysplasia congenita,0.0004140565870329,0,[94068]
+UDN-P9,"spondyloepiphyseal dysplasia tarda, X-linked",0.0004133584443479,0,10737
+UDN-P9,"spondylometaphyseal dysplasia, 'corner fracture' type",0.0004127403371967,0,[93315]
+UDN-P9,pseudoachondroplasia,0.0004125143168494,0,[750]
+UDN-P9,Coxa vara,0.000412239722209,0,Coxa vara
+UDN-P8,combined oxidative phosphorylation deficiency 39,0.0003771049669012,0,[565624]
+UDN-P8,"leukodystrophy, hypomyelinating, 20",0.0003728091833181,0,33657
+UDN-P8,pyruvate dehydrogenase E3-binding protein deficiency,0.0003713524783961,0,[255182]
+UDN-P8,intellectual disability-epilepsy-extrapyramidal syndrome,0.000369677611161,0,[468620]
+UDN-P8,combined oxidative phosphorylation defect type 27,0.0003694282495416,0,[477774]
+UDN-P8,severe Canavan disease,0.0003670282894745,0,[314911]
+UDN-P8,guanidinoacetate methyltransferase deficiency,0.0003666807315312,0,[382]
+UDN-P8,"neurodevelopmental disorder, mitochondrial, with abnormal movements and lactic acidosis, with or without seizures",0.0003646771074272,0,[572798]
+UDN-P8,pyruvate dehydrogenase phosphatase deficiency,0.0003641054208856,0,[79246]
+UDN-P8,childhood-onset motor and cognitive regression syndrome with extrapyramidal movement disorder,0.00036309598363,0,[500180]
+UDN-P2,methylmalonic aciduria and homocystinuria type cblF,0.0003374280931893,0,[79284]
+UDN-P2,neonatal hemochromatosis,0.0003332369087729,0,[446]
+UDN-P2,homozygous 11P15-p14 deletion syndrome,0.0003293412737548,0,11678
+UDN-P2,ALG8-CDG,0.000328178924974,0,[79325]
+UDN-P2,congenital anemia,0.000324971275404,0,577
+UDN-P2,familial thyroid dyshormonogenesis 1,0.0003192963486071,0,20716
+UDN-P2,congenital hypothyroidism due to maternal intake of antithyroid drugs,0.0003163111105095,0,[226313]
+UDN-P2,"methylmalonic acidemia with homocystinuria, type cblJ",0.000315255660098,0,[369955]
+UDN-P2,COG6-CGD,0.0003150861884932,0,[464443]
+UDN-P2,congenital toxoplasmosis,0.0003135850711259,0,[858]

dx_characterization_scores_attn.csv

@@ -0,0 +1,55 @@
+patient_id,phenotypes,degrees,attention
+UDN-P2,Abnormality of vision,111,0.0989853367209434
+UDN-P2,Global developmental delay,1330,0.1393505930900573
+UDN-P2,Abnormality of the liver,65,0.0837447270750999
+UDN-P2,Failure to thrive in infancy,1,0.1541623473167419
+UDN-P2,Exocrine pancreatic insufficiency,4,0.0962460488080978
+UDN-P2,Abdominal pain,257,0.0696877762675285
+UDN-P2,Asthma,177,0.0721269026398658
+UDN-P2,Duodenal atresia,34,0.0989027693867683
+UDN-P2,Intestinal malrotation,91,0.0801789388060569
+UDN-P2,Gastroparesis,23,0.106614664196968
+UDN-P9,Flat glenoid fossa,1,0.0665825009346008
+UDN-P9,Abnormality of the vertebral column,76,0.0361983776092529
+UDN-P9,Platyspondyly,129,0.0615288130939006
+UDN-P9,Cutis marmorata,62,0.0318356454372406
+UDN-P9,Wormian bones,1,0.0537327453494071
+UDN-P9,Scoliosis,801,0.0558811277151107
+UDN-P9,Spondylometaphyseal dysplasia,9,0.0940656214952468
+UDN-P9,Coxa vara,56,0.0579654388129711
+UDN-P9,Abnormality of the lower limb,48,0.0458225421607494
+UDN-P9,Thoracic scoliosis,30,0.0464530810713768
+UDN-P9,Genu varum,57,0.073094867169857
+UDN-P9,Tibial bowing,43,0.0525600053369998
+UDN-P9,Abnormal form of the vertebral bodies,138,0.0546819157898426
+UDN-P9,Short stature,1167,0.0585363544523716
+UDN-P9,Abnormality of the femoral metaphysis,10,0.0352263785898685
+UDN-P9,Abnormality of femoral epiphysis,10,0.0409332402050495
+UDN-P9,Abnormality of lower limb epiphysis morphology,6,0.0124136246740818
+UDN-P9,Acetabular dysplasia,1,0.0529508516192436
+UDN-P9,Short femoral neck,1,0.0695368871092796
+UDN-P7,Elevated serum creatine kinase,219,0.156127393245697
+UDN-P7,Generalized muscle weakness,1,0.139058843255043
+UDN-P7,Muscular dystrophy,52,0.1362844258546829
+UDN-P7,Proximal muscle weakness,120,0.1536363512277603
+UDN-P7,Calf muscle pseudohypertrophy,1,0.1570710241794586
+UDN-P7,Abnormal muscle fiber morphology,39,0.0849768370389938
+UDN-P7,Toe walking,1,0.1728451550006866
+UDN-P8,Muscle weakness,392,0.0540948249399662
+UDN-P8,Hypertrophic cardiomyopathy,153,0.0568107031285762
+UDN-P8,Cerebral atrophy,231,0.0622156262397766
+UDN-P8,Hypoplasia of the corpus callosum,332,0.0678444355726242
+UDN-P8,Limb hypertonia,47,0.0716666877269744
+UDN-P8,Lactic acidosis,122,0.0437235236167907
+UDN-P8,Abnormality of temperature regulation,23,0.0598855391144752
+UDN-P8,Respiratory failure requiring assisted ventilation,31,0.0432068929076194
+UDN-P8,Cerebral hypomyelination,19,0.0751257538795471
+UDN-P8,Infantile axial hypotonia,2,0.0626801550388336
+UDN-P8,EEG with focal slow activity,4,0.0187200624495744
+UDN-P8,EEG with generalized slow activity,13,0.0373874828219413
+UDN-P8,Infantile spasms,3,0.0651202574372291
+UDN-P8,Elevated brain lactate level by MRS,4,0.0452156253159046
+UDN-P8,Profound global developmental delay,1,0.0835469588637352
+UDN-P8,Abnormal cerebral morphology,25,0.0396268218755722
+UDN-P8,Cerebral visual impairment,97,0.0617175139486789
+UDN-P8,Oral-pharyngeal dysphagia,1,0.0514112450182437

exomiser_gene_discovery_scores.csv

@@ -0,0 +1,21 @@
+patient_id,genes,similarities,correct_gene_label
+UDN-P1,UBE3A,0.77525794506073,0
+UDN-P1,POLR3A,0.7635941505432129,1
+UDN-P2,AK2,0.7269436120986938,0
+UDN-P2,INSR,0.7229311466217041,0
+UDN-P2,SMC1A,0.7217983603477478,0
+UDN-P1,KMT2E,0.705829381942749,0
+UDN-P1,TNIK,0.6911647319793701,0
+UDN-P1,ORC4,0.683530867099762,0
+UDN-P2,UBE3A,0.6692474484443665,0
+UDN-P1,CTU2,0.6311229467391968,0
+UDN-P1,TGIF1,0.561279833316803,0
+UDN-P2,NKX2-3,0.5290572643280029,0
+UDN-P2,SSTR2,0.524275541305542,0
+UDN-P2,PABPC1,0.5164581537246704,0
+UDN-P1,TBP,0.5104368329048157,0
+UDN-P1,MED16,0.5011780858039856,0
+UDN-P1,DVL3,0.4911206364631653,0
+UDN-P2,NBPF1,0.4773442149162292,0
+UDN-P2,GLYR1,0.4720344543457031,1
+UDN-P2,PBRM1,0.4607828855514526,0

exomiser_gene_discovery_scores_attn.csv

@@ -0,0 +1,57 @@
+patient_id,phenotypes,degrees,attention
+UDN-P2,Abnormality of vision,111,0.1183807775378227
+UDN-P2,Global developmental delay,1330,0.1499209105968475
+UDN-P2,Abnormality of the liver,65,0.0950504913926124
+UDN-P2,Failure to thrive in infancy,1,0.1553362309932708
+UDN-P2,Exocrine pancreatic insufficiency,4,0.0865935012698173
+UDN-P2,Abdominal pain,257,0.068713754415512
+UDN-P2,Asthma,177,0.0740201994776725
+UDN-P2,Duodenal atresia,34,0.064612977206707
+UDN-P2,Intestinal malrotation,91,0.0643364712595939
+UDN-P2,Gastroparesis,23,0.1230346783995628
+UDN-P1,Thick lower lip vermilion,1,0.0236996822059154
+UDN-P1,Hydrocephalus,304,0.0230413302779197
+UDN-P1,Microcephaly,1006,0.0316418893635273
+UDN-P1,Low anterior hairline,4,0.0270521882921457
+UDN-P1,Mandibular prognathia,6,0.026597810909152
+UDN-P1,Alacrima,1,0.0138036329299211
+UDN-P1,Thick eyebrow,8,0.0170831736177206
+UDN-P1,Exotropia,94,0.0167111363261938
+UDN-P1,Synophrys,3,0.0241384990513324
+UDN-P1,Delayed eruption of teeth,131,0.0224792174994945
+UDN-P1,Abnormal sternum morphology,51,0.0120808342471718
+UDN-P1,Hypertension,278,0.0193403456360101
+UDN-P1,Hypotonia,911,0.0355296544730663
+UDN-P1,Global developmental delay,1330,0.0301800202578306
+UDN-P1,Dystonia,360,0.0236831586807966
+UDN-P1,Absent speech,1,0.0214502979069948
+UDN-P1,Failure to thrive,648,0.0351606905460357
+UDN-P1,Growth delay,435,0.0286059584468603
+UDN-P1,Pancreatitis,63,0.0134885320439934
+UDN-P1,Overlapping toe,2,0.0212031248956918
+UDN-P1,Constipation,270,0.0162120759487152
+UDN-P1,Gastroesophageal reflux,268,0.0238284692168235
+UDN-P1,Cerebral atrophy,231,0.0270951297134161
+UDN-P1,Hypoplasia of the corpus callosum,332,0.0298781190067529
+UDN-P1,Respiratory insufficiency,325,0.0320345684885978
+UDN-P1,Ventriculomegaly,409,0.0279995389282703
+UDN-P1,Coarse hair,2,0.032378576695919
+UDN-P1,Drooling,85,0.0170378927141428
+UDN-P1,Febrile seizure (within the age range of 3 months to 6 years),62,0.0118239596486091
+UDN-P1,Developmental regression,203,0.0243585351854562
+UDN-P1,Abnormal cerebral white matter morphology,76,0.0176227763295173
+UDN-P1,Short stature,1167,0.0382893718779087
+UDN-P1,Chronic pancreatitis,1,0.0170067008584737
+UDN-P1,Premature loss of primary teeth,1,0.0167370326817035
+UDN-P1,Premature loss of teeth,32,0.0260231960564851
+UDN-P1,Chronic lung disease,1,0.0176592171192169
+UDN-P1,Moderate sensorineural hearing impairment,1,0.009951589629054
+UDN-P1,Laryngeal cleft,1,0.0077759157866239
+UDN-P1,Laryngeal calcification,1,0.0125970430672168
+UDN-P1,Low hanging columella,2,0.0163100585341453
+UDN-P1,Prominent eyelashes,1,0.024880614131689
+UDN-P1,Gastrostomy tube feeding in infancy,1,0.0276554562151432
+UDN-P1,Midface retrusion,1,0.0258742850273847
+UDN-P1,Premature adrenarche,1,0.0266624484211206
+UDN-P1,T2 hypointense brainstem,1,0.0004453512665349
+UDN-P1,T2 hypointense basal ganglia,49,0.0048909387551248

gene_discovery_scores.csv

@@ -0,0 +1,23 @@
+patient_id,genes,similarities,correct_gene_label
+UDN-P2,CTH,0.3697766065597534,0
+UDN-P2,WASHC2A,0.1429679691791534,0
+UDN-P2,GLYR1,0.4873602986335754,1
+UDN-P2,PRKACA,0.2901750802993774,0
+UDN-P2,SMPD3,0.0044486671686172,0
+UDN-P1,ANO3,0.1011797934770584,0
+UDN-P1,TYMP,-0.1458933055400848,0
+UDN-P1,ORC4,0.6925094127655029,0
+UDN-P1,DST,-0.0270717144012451,0
+UDN-P1,NDUFAF5,0.1227701157331466,0
+UDN-P1,WDFY4,0.5533968806266785,0
+UDN-P1,TOPORS,-0.1653318554162979,0
+UDN-P1,SLK,-0.1769217252731323,0
+UDN-P1,GMPPA,0.2679024338722229,0
+UDN-P1,APC,0.2313176691532135,0
+UDN-P1,INSL3,0.0432731509208679,0
+UDN-P1,ZFYVE26,0.3379367589950561,0
+UDN-P1,KAT6A,0.7698580026626587,0
+UDN-P1,POLR3A,0.7476195693016052,1
+UDN-P1,DYNAP,-0.2454878091812133,0
+UDN-P1,NCOR2,0.2544351518154144,0
+UDN-P1,PIWIL3,-0.2566407322883606,0

gene_discovery_scores_attn.csv

@@ -0,0 +1,57 @@
+patient_id,phenotypes,degrees,attention
+UDN-P2,Abnormality of vision,111,0.1178910732269287
+UDN-P2,Global developmental delay,1330,0.1363666206598281
+UDN-P2,Abnormality of the liver,65,0.0964500233530998
+UDN-P2,Failure to thrive in infancy,1,0.1844952553510666
+UDN-P2,Exocrine pancreatic insufficiency,4,0.0708952322602272
+UDN-P2,Abdominal pain,257,0.0660589486360549
+UDN-P2,Asthma,177,0.0678283423185348
+UDN-P2,Duodenal atresia,34,0.0732858255505561
+UDN-P2,Intestinal malrotation,91,0.0733114182949066
+UDN-P2,Gastroparesis,23,0.1134173870086669
+UDN-P1,Thick lower lip vermilion,1,0.0254656299948692
+UDN-P1,Hydrocephalus,304,0.0226140450686216
+UDN-P1,Microcephaly,1006,0.0317770205438137
+UDN-P1,Low anterior hairline,4,0.0236586183309555
+UDN-P1,Mandibular prognathia,6,0.0236049313098192
+UDN-P1,Alacrima,1,0.0144500378519296
+UDN-P1,Thick eyebrow,8,0.0233250595629215
+UDN-P1,Exotropia,94,0.0174672622233629
+UDN-P1,Synophrys,3,0.0241911858320236
+UDN-P1,Delayed eruption of teeth,131,0.0234366580843925
+UDN-P1,Abnormal sternum morphology,51,0.0120793534442782
+UDN-P1,Hypertension,278,0.0188926625996828
+UDN-P1,Hypotonia,911,0.03265380859375
+UDN-P1,Global developmental delay,1330,0.027355071157217
+UDN-P1,Dystonia,360,0.0240448117256164
+UDN-P1,Absent speech,1,0.022760335355997
+UDN-P1,Failure to thrive,648,0.0337891578674316
+UDN-P1,Growth delay,435,0.0274257939308881
+UDN-P1,Pancreatitis,63,0.0125568015500903
+UDN-P1,Overlapping toe,2,0.0273327771574258
+UDN-P1,Constipation,270,0.0161661840975284
+UDN-P1,Gastroesophageal reflux,268,0.0214648023247718
+UDN-P1,Cerebral atrophy,231,0.0250347331166267
+UDN-P1,Hypoplasia of the corpus callosum,332,0.0269252881407737
+UDN-P1,Respiratory insufficiency,325,0.0316686779260635
+UDN-P1,Ventriculomegaly,409,0.0280382838100194
+UDN-P1,Coarse hair,2,0.024032786488533
+UDN-P1,Drooling,85,0.0192697960883378
+UDN-P1,Febrile seizure (within the age range of 3 months to 6 years),62,0.0105393650010228
+UDN-P1,Developmental regression,203,0.0249004419893026
+UDN-P1,Abnormal cerebral white matter morphology,76,0.0189827494323253
+UDN-P1,Short stature,1167,0.036681056022644
+UDN-P1,Chronic pancreatitis,1,0.0087832231074571
+UDN-P1,Premature loss of primary teeth,1,0.0137059595435857
+UDN-P1,Premature loss of teeth,32,0.0200045183300972
+UDN-P1,Chronic lung disease,1,0.030693894252181
+UDN-P1,Moderate sensorineural hearing impairment,1,0.0128651773557066
+UDN-P1,Laryngeal cleft,1,0.0062378644943237
+UDN-P1,Laryngeal calcification,1,0.0187820736318826
+UDN-P1,Low hanging columella,2,0.0193525701761245
+UDN-P1,Prominent eyelashes,1,0.0317739509046077
+UDN-P1,Gastrostomy tube feeding in infancy,1,0.0312498826533556
+UDN-P1,Midface retrusion,1,0.0272352565079927
+UDN-P1,Premature adrenarche,1,0.0155880162492394
+UDN-P1,T2 hypointense brainstem,1,0.0003417030384298
+UDN-P1,T2 hypointense basal ganglia,49,0.0108007565140724

patients_like_me_scores.csv

@@ -0,0 +1,41 @@
+patient_id,candidate_patients,similarities,correct_label,genes,diseases,disease_ids
+UDN-P4,SIM-P7513,0.0001346864301012,0.0,REEP1,Autosomal dominant spastic paraplegia type 31,101011
+UDN-P4,SIM-P38172,0.0001341508032055,0.0,KIF1A,Autosomal spastic paraplegia type 30,101010
+UDN-P4,SIM-P31298,0.0001340870658168,0.0,DDHD1,Autosomal recessive spastic paraplegia type 28,101008
+UDN-P4,SIM-P19068,0.0001340282615274,1.0,CAPN1,Autosomal recessive spastic paraplegia type 76,488594
+UDN-P4,SIM-P19513,0.0001340070302831,0.0,ATL1,Autosomal dominant spastic paraplegia type 3,100984
+UDN-P4,SIM-P36075,0.000133997775265,0.0,MTPAP,Autosomal recessive spastic ataxia-optic atrophy-dysarthria syndrome,254343
+UDN-P4,SIM-P18901,0.000133881723741,0.0,TDP1,Spinocerebellar ataxia with axonal neuropathy type 1,94124
+UDN-P4,SIM-P20117,0.000133869878482,0.0,SPAST,Autosomal dominant spastic paraplegia type 4,100985
+UDN-P4,SIM-P41708,0.0001337969733867,0.0,SPG19,Autosomal dominant spastic paraplegia type 19,100999
+UDN-P4,SIM-P24993,0.0001337462454102,0.0,SPG19,Autosomal dominant spastic paraplegia type 19,100999
+UDN-P6,SIM-P34720,0.0001335124979959,0.0,PMM2,PMM2-CDG,79318
+UDN-P6,SIM-P6961,0.000133484849357,0.0,AP1S2,X-linked intellectual disability-hypotonia-facial dysmorphism-aggressive behavior syndrome,85329
+UDN-P6,SIM-P39957,0.0001328073558397,0.0,TCF4,Pitt-Hopkins syndrome,2896
+UDN-P6,SIM-P13143,0.0001326532074017,0.0,SSR4,SSR4-CDG,370927
+UDN-P6,SIM-P21183,0.0001325271441601,0.0,KAT6A,Autosomal dominant intellectual disability-craniofacial anomalies-cardiac defects syndrome,457193
+UDN-P6,SIM-P27029,0.0001323336182394,0.0,ATRX,Chudley-Lowry-Hoar syndrome,93971
+UDN-P6,SIM-P26917,0.0001322419411735,0.0,NONO,Macrocephaly-intellectual disability-left ventricular non compaction syndrome,466791
+UDN-P6,SIM-P13788,0.0001321789313806,0.0,NRXN1,Pitt-Hopkins-like syndrome,221150
+UDN-P6,SIM-P3592,0.0001321435265708,0.0,FRMD4A,Severe intellectual disability-corpus callosum agenesis-facial dysmorphism-cerebellar ataxia syndrome,466688
+UDN-P6,SIM-P24214,0.0001321362215094,0.0,AP1S2,X-linked intellectual disability-hypotonia-facial dysmorphism-aggressive behavior syndrome,85329
+UDN-P5,SIM-P12152,0.0001249182969331,0.0,MYD88,Waldenström macroglobulinemia,33226
+UDN-P5,SIM-P23023,0.0001241921563632,0.0,TNFRSF1A,Tumor necrosis factor receptor 1 associated periodic syndrome,32960
+UDN-P5,SIM-P30461,0.0001231428614119,0.0,MYD88,Waldenström macroglobulinemia,33226
+UDN-P5,SIM-P6717,0.0001229638874065,0.0,MYD88,Waldenström macroglobulinemia,33226
+UDN-P5,SIM-P30356,0.00012291384337,0.0,PSTPIP1,Pyogenic arthritis-pyoderma gangrenosum-acne syndrome,69126
+UDN-P5,SIM-P35189,0.0001224889274453,0.0,NLRC4,Familial cold urticaria,47045
+UDN-P5,SIM-P20277,0.0001221649872604,0.0,CPOX,Hereditary coproporphyria,79273
+UDN-P5,SIM-P7603,0.000121750497783,0.0,NLRP3,Familial cold urticaria,47045
+UDN-P5,SIM-P8138,0.0001210815316881,0.0,MEFV,Familial Mediterranean fever,342
+UDN-P5,SIM-P36107,0.0001210761329275,0.0,TNFRSF1A,Tumor necrosis factor receptor 1 associated periodic syndrome,32960
+UDN-P3,SIM-P17696,0.0001152271579485,0.0,GRIA3,X-linked intellectual disability due to GRIA3 mutations,364028
+UDN-P3,SIM-P12018,0.0001116794228437,1.0,RPS6KA3,Symptomatic form of Coffin-Lowry syndrome in female carriers,276630
+UDN-P3,SIM-P13072,0.0001079059220501,0.0,THOC2,X-linked intellectual disability-short stature-overweight syndrome,457240
+UDN-P3,SIM-P27682,0.0001075164036592,0.0,AP1S2,Fried syndrome,85335
+UDN-P3,SIM-P17495,0.0001061590883182,0.0,SMS,"X-linked intellectual disability, Snyder type",3063
+UDN-P3,SIM-P12040,0.000106000341475,0.0,AP1S2,Fried syndrome,85335
+UDN-P3,SIM-P38032,0.0001052861043717,0.0,UBE3A,Angelman syndrome due to a point mutation,411511
+UDN-P3,SIM-P15144,0.0001048136182362,0.0,PHF8,"X-linked intellectual disability, Siderius type",85287
+UDN-P3,SIM-P2560,0.0001046940378728,0.0,AP1S2,Fried syndrome,85335
+UDN-P3,SIM-P13027,0.000104646103864,0.0,FMR1,Fragile X syndrome,908

patients_like_me_scores_attn.csv

@@ -0,0 +1,84 @@
+patient_id,phenotypes,degrees,attention
+UDN-P4,Nystagmus,554,0.052525907754898
+UDN-P4,Horizontal nystagmus,56,0.0585071220993995
+UDN-P4,Spasticity,411,0.0557427555322647
+UDN-P4,Dysarthria,369,0.0397082976996898
+UDN-P4,Brisk reflexes,1,0.0718949064612388
+UDN-P4,Headache,257,0.0289554204791784
+UDN-P4,Unsteady gait,103,0.0579096935689449
+UDN-P4,Limb dysmetria,9,0.0552756935358047
+UDN-P4,Upper motor neuron dysfunction,44,0.0472515001893043
+UDN-P4,Steppage gait,32,0.0555065162479877
+UDN-P4,Paresthesia,116,0.0359307341277599
+UDN-P4,EMG abnormality,99,0.0366697795689106
+UDN-P4,Proportionate short stature,29,0.0580268092453479
+UDN-P4,Lumbar spinal canal stenosis,1,0.0060006459243595
+UDN-P4,Impaired distal vibration sensation,13,0.0658590272068977
+UDN-P4,Abnormal pyramidal sign,147,0.0405740477144718
+UDN-P4,Cervical spondylosis,3,0.042157094925642
+UDN-P4,Spondylolisthesis at L5-S1,1,0.0032220152206718
+UDN-P4,Distal lower limb amyotrophy,23,0.0685843080282211
+UDN-P4,Weakness due to upper motor neuron dysfunction,10,0.029885521158576
+UDN-P4,Ankle clonus,1,0.0551076084375381
+UDN-P4,Brain atrophy,65,0.0347045883536338
+UDN-P3,Low anterior hairline,4,0.0839744806289672
+UDN-P3,Psychosis,91,0.0314286798238754
+UDN-P3,Oligomenorrhea,29,0.0356979817152023
+UDN-P3,Tapered finger,1,0.0815264210104942
+UDN-P3,Mental deterioration,127,0.0507740229368209
+UDN-P3,Joint laxity,166,0.0528562292456626
+UDN-P3,Pes planus,224,0.0508626289665699
+UDN-P3,Sleep disturbance,191,0.0410025417804718
+UDN-P3,Abnormal brainstem morphology,35,0.055194191634655
+UDN-P3,Abnormal rapid eye movement sleep,6,0.0391974486410617
+UDN-P3,Scoliosis,801,0.0617553479969501
+UDN-P3,"High, narrow palate",4,0.0490836761891841
+UDN-P3,Short nose,3,0.0597070343792438
+UDN-P3,Abnormality of the clivus,3,0.0444605834782123
+UDN-P3,Prominent forehead,1,0.0760357454419136
+UDN-P3,Abnormal occipital bone morphology,1,0.031018890440464
+UDN-P3,Abnormal upper to lower segment ratio,3,0.0362525805830955
+UDN-P3,Increased head circumference,14,0.0620610117912292
+UDN-P3,Low levels of vitamin D,1,0.0350590944290161
+UDN-P3,Thyroiditis,31,0.0220514684915542
+UDN-P5,Cutis marmorata,62,0.0696348845958709
+UDN-P5,Skin rash,146,0.0380239747464656
+UDN-P5,Muscle weakness,392,0.0529503189027309
+UDN-P5,Joint hypermobility,142,0.1087778359651565
+UDN-P5,Exocrine pancreatic insufficiency,4,0.0637081339955329
+UDN-P5,Weight loss,253,0.0690883025527
+UDN-P5,Recurrent fever,66,0.0378629937767982
+UDN-P5,Constipation,270,0.0589890293776989
+UDN-P5,Migraine,101,0.0587076544761657
+UDN-P5,Episodic abdominal pain,1,0.0709817335009574
+UDN-P5,Recurrent infections,214,0.082623191177845
+UDN-P5,Arthralgia,217,0.0823353752493858
+UDN-P5,Easy fatigability,2,0.0924237966537475
+UDN-P5,Gastritis,13,0.0227577649056911
+UDN-P5,Peripheral neuropathy,215,0.0911349803209304
+UDN-P6,Urinary incontinence,101,0.0282541252672672
+UDN-P6,Narrow mouth,1,0.0286488384008407
+UDN-P6,Thin upper lip vermilion,222,0.0543260090053081
+UDN-P6,Macrocephaly,355,0.0478888861835002
+UDN-P6,Hypertelorism,755,0.0466706231236457
+UDN-P6,Smooth philtrum,1,0.0513933710753917
+UDN-P6,Low-set ears,604,0.0431698933243751
+UDN-P6,Exotropia,94,0.0236861575394868
+UDN-P6,Global developmental delay,1330,0.0373244658112525
+UDN-P6,Generalized hypotonia,762,0.0430077537894248
+UDN-P6,Absent speech,1,0.0480412095785141
+UDN-P6,Failure to thrive,648,0.0439043194055557
+UDN-P6,Growth delay,435,0.0379057824611663
+UDN-P6,Gastroesophageal reflux,268,0.0282815545797348
+UDN-P6,Chronic diarrhea,9,0.0330403670668602
+UDN-P6,Myopathic facies,2,0.0682374909520149
+UDN-P6,Broad-based gait,3,0.0389422886073589
+UDN-P6,Gait imbalance,6,0.0347393229603767
+UDN-P6,Bowel incontinence,51,0.0291043147444725
+UDN-P6,Delayed skeletal maturation,290,0.0552288256585598
+UDN-P6,Anisocoria,16,0.0166235510259866
+UDN-P6,Monocular strabismus,1,0.0136106135323643
+UDN-P6,Prominent forehead,1,0.0529596321284771
+UDN-P6,Chronic constipation,3,0.0185318272560834
+UDN-P6,Small pituitary gland,11,0.0520597584545612
+UDN-P6,Low levels of vitamin D,1,0.0244190022349357

requirements.txt

@@ -0,0 +1,3 @@
+pandas
+gradio
+