Initial commit: digital-patients pipeline (clean, no large files)

Large reference/model files excluded from repo - to be staged to S3 or baked into Docker images.

test_gen_patient.nf

@@ -0,0 +1,125 @@
+nextflow.enable.dsl=2
+
+//SYNTHEA
+params.n_pat = 2 //10 //number of patients to generate, must be => 2 (one male one female)
+params.percent_male = 0.5 // between 0-1, percent of male patients
+params.age = '18-80' //age range of the population, expressed as age-age
+// params.state = '"District of Columbia"' //'Hawaii' //'Texas' //'Georgia'
+// params.city = '' //'Washington' //'' //'Honolulu' //'Houston' //'Atlanta'
+params.imputed_store = '/mnt/Avatar/imputed/ukbb/imputed' //'/rosalind/ukbb/imputed'
+params.pheno_store = '/mnt/Avatar/dd/synthea/metadata/ukbb_phenotypes_filtered.csv'
+
+// Keep both versions of disease params
+params.disease = 'schizophrenia' //'schizophrenia' //'healthy' //'leukaemia' //'leukaemia' //'Purpura and other haemorrhagic conditions' //['tongue cancer', 'dementia', 'arthritis'] //'tongue cancer' //'dementia'
+params.healthy_dir = '/Workspace/next/registry/pipelines/digital_patient' //'/mnt/Avatar/digital_patient' // data copy is here 
+
+//params.n_var = 100
+params.outdir = '/mnt/OmicNAS/dd/digital_patient'
+
+//BORZOI
+params.container_borzoi = 'borzoi:latest'
+params.containerOptions = '--gpus all --rm -v /mnt:/mnt'
+params.project_name = 'test'
+params.mane = '/Workspace/next/registry/pipelines/digital_patient/MANE.GRCh38.v1.3.update.tsv'
+
+//VCF2PROT
+params.container_vcf2prot = "vcf2prot:latest"
+
+//RNA2PROTEINEXPRESSION
+params.container_rna2protexpression = 'rna2protexpression:latest'
+params.containerOptions_rna2protexpression = '--gpus all --rm -v /mnt:/mnt -v /dbs:/dbs'
+
+//CORTO
+params.container_corto = 'corto:latest'
+params.regulon = '/Workspace/next/registry/pipelines/digital_patient/regulon.rda'
+
+//CIBERSORT
+params.cibersortx_username = "gabriel.richman.2009@anderson.ucla.edu"
+params.cibersortx_token = "b5b39e563fb947df4cfd4843d40fdb99"
+params.container_ecotyper = 'ecotyper:latest'
+params.signature_matrix = "/Workspace/next/registry/pipelines/digital_patient/LM22_sourceGEP_ensg.txt"
+params.ecotyper_outdir = "/mnt/OmicNAS/olamide/ecotyper/results/ecotyper"
+
+//SYNTHEA
+include {generate_patients} from './main_synthea.nf'
+// include {get_disease_stats_no_patients} from './main_synthea.nf'
+// include {generate_m_variants_cudf} from './main_synthea.nf'
+// include {generate_f_variants_cudf} from './main_synthea.nf'
+// include {make_vcfs} from './main_synthea.nf'
+// include {generate_m_healthy_cudf} from './main_synthea.nf'
+// include {generate_f_healthy_cudf} from './main_synthea.nf'
+
+// //BORZOI
+// include {FILTER_VCF} from './main_borzoi.nf'
+// include {PREDICT_EXPRESSION} from './main_borzoi.nf'
+// include {CREATE_PROTEIN_CLUSTER} from './main_borzoi.nf'
+
+// //VCF2PROT
+// include {VCF2PROT} from './main_vcf2prot.nf'
+
+// //RNA2PROTEINEXPRESSION
+// include {RNA2PROTEXPRESSION} from './main_rna2proteinexpression'
+
+// //CORTO
+// include {CORTO} from './main_corto.nf'
+
+// //CIBERSORT
+// include {CONVERT_TO_TXT} from './main_cibersortx.nf'
+// include {CIBERSORTx_FRACTIONS} from './main_cibersortx.nf'
+// include {CIBERSORTx_HIRES} from './main_cibersortx.nf'
+// include {ADD_TISSUE_NAMES_TO_CIBERSORTX} from './main_cibersortx.nf'
+
+workflow {
+  generate_patients()
+
+  // //SYNTHEA
+  // switch (params.disease) {
+  //   case 'healthy':
+  //     //healthy 
+  //     generate_m_healthy_cudf(params.healthy_dir)
+  //     generate_f_healthy_cudf(params.healthy_dir)
+  //     m_healthy = generate_m_healthy_cudf.out
+  //     f_healthy = generate_f_healthy_cudf.out
+  //     txt_ch = f_healthy.mix(m_healthy).flatten()
+  //     break
+  //   default:
+  //     //disease
+  //     get_disease_stats_no_patients()
+  //     generate_m_variants_cudf(get_disease_stats_no_patients.out)
+  //     generate_f_variants_cudf(get_disease_stats_no_patients.out)
+  //     f_var = generate_f_variants_cudf.out
+  //     m_var = generate_m_variants_cudf.out
+  //     txt_ch = f_var.mix(m_var).flatten()
+  // }
+  // make_vcfs(txt_ch)
+
+  // //BORZOI
+  // // Prepare the VCF file(s)
+  // //vcf_ch = Channel.fromPath(make_vcfs.out)
+  // // THIS IS USES TO CREATE PROTEIN CLUSTER ON A SAME BORZOI INPUT AND SPLIT ONES TOO BIG FOR BORZOI (LEN = 524288)
+  // // THIS STEP NEEDS TO BE RUN TO CREATE DATA FOR NEXT STEP ONLY IF NEW REFERENCE/RNAS ARE GOING TO BE USED
+  // //CREATE_PROTEIN_CLUSTER(params.mane)
+  // FILTER_VCF(params.mane, make_vcfs.out)
+  // PREDICT_EXPRESSION(FILTER_VCF.out, params.mane)
+
+  // //VCF2PROT
+  // VCF2PROT(make_vcfs.out, FILTER_VCF.out)
+
+  // //RNA2PROTEINEXPRESSION
+  // // Updated RNA to protein expression step to run only a single process at a time, avoids CUDA out-of-memory errors.
+  // PREDICT_EXPRESSION.out
+  //     .collect()
+  //     .flatten()
+  //     .set { rna_input }
+  // RNA2PROTEXPRESSION(rna_input)
+
+  // //CORTO
+  // CORTO(PREDICT_EXPRESSION.out, params.regulon)
+
+  // //CIBERSORT
+  // signature_file = Channel.fromPath(params.signature_matrix, checkIfExists: true)
+  // CONVERT_TO_TXT(PREDICT_EXPRESSION.out)
+  // CIBERSORTx_FRACTIONS(CONVERT_TO_TXT.out, signature_file)
+  // CIBERSORTx_HIRES(CONVERT_TO_TXT.out, CIBERSORTx_FRACTIONS.out, signature_file)
+  // ADD_TISSUE_NAMES_TO_CIBERSORTX(CONVERT_TO_TXT.out, CIBERSORTx_HIRES.out)
+}