5iL/ddlmZddlZddlZddlZddlmZddlmZdZ dZ dZ d Z d Z dd Zd Zd ZdZdZdZdZdZdZdZddZ ddZ ddZy)) annotationsN)extract_utils_basepath)GENOTYPE_TYPES)refno-callz#SELECT\s+(DISTINCT\s+)?(.+?)\s+FROMz((`?(\w+)?`?\.)?`?(\w+)`?( AS `?(\w+)`?)?cp|jdxst}|Dcgc] }|tvs |c}Scc}w)zl Returns a list of genotype types for the filter that are use in genotype and allele table queries. genotype_type)getrGENOTYPE_ONLY_TYPES) filter_dictgenotype_typesr s /home/marpiech/ifpan-abm-pgxpred/analysis/marpiech-gwas-test/venv/lib/python3.12/site-packages/dxpy/dx_extract_utils/germline_utils.pyget_genotype_typesrs3 !___5GN/= jmViAiM jj js 33c|||d}|jdxs t|}g}|jD]!\}}||vs ||r|j|#|S)zf Returns a list of genotype types for the filter that are use in genotype table only queries. )rhalfrr )r listitemsappend) r exclude_refdataexclude_halfrefexclude_nocallgenotype_type_exclude_flag_maprgenotype_only_typesr excludeds rget_genotype_only_typesrsx !&" !___5]>\9]N#A#G#G#I6 x N *x/CH  & &} 56 c<|rdDcgc]}||vr| c}SgScc}w)z Infer option require all genotypes types to be queried. If users wishes to obtain only certain types of genotypes, reminder of the types should be filtered out post querying )rzhet-refhomzhet-altrr)requested_typestypes r%get_types_to_filter_out_when_inferingr#/s< L    &  !! ! s cnd|dvr.|jdr|jd|d<|d|d<yy)N CohortBrowser recordTypesbaseSqlbase_sqlfilters)r )resppayloads radd_germline_base_sqlr,>sA$}-- 88I "&((9"5GJ !)_ .rc:d|vr|dr|djd\}}}}n%d|vr!|dr|djddd\}}d}|jdd}jrt|dt|fSt d|t|fS)N allele_id_locus_id sample_idinf)splitr isdigitintfloat)dchromposr/altr3s rsort_germline_variantr=EsaAkN{^11#6sAs qQz]z]((-bq1 sk2&I }}5z2s3xi77 <C#y 88rc#<Ktjt|jd}|j dDcgc]}|j c}D]:}tjt |j}|d|d|df<ycc}ww)zz Parses the SELECT list of a SQL statement and returns a generator of named expressions table, column, and alias. r1,rN)rematchSELECT_LIST_REGEXgroupr5stripNAMED_EXPRESSION_REGEXgroups)sqlselect_list_matchxnamed_expressionnamed_expression_matchs r"_parse_sql_select_named_expressionrMQs!2C8>>qA0A0G0G0LM1QWWYM^!#*@BR!S!Z!Z!\$Q')?)BDZ[\D]]]^Ms=BBABc Xt|Dcic] \}}}|||f }}}}g}|D]}t|d}||vr,||\}}|jdj|||A||vr/|jdj||d||dt|jdj|d j |} t j t|jd} t jtd j| | |Scc}}}w) z Harmonizes the SELECT list of a SQL statement to include all columns in return_columns. NULL values are used for columns not in the SELECT list. rz)`{table}`.`{column}` AS `{return_column}`)tablecolumn return_columnz`{table}`.`{column}` AS `ref`r)rOrPzNULL AS `{return_column}`)rQz, z#SELECT {distinct}{select_list} FROM)distinct select_list) rMtuplerformatjoinrArBrCrDsub) rHreturn_columnskwargsrOrPalias select_info select_listsrQrSrRs r&_harmonize_sql_select_named_expressionr][sb GiilFmnn.BeVU55&/)nKnL' a m,Q/ K '' 6ME6    K R RY^Z`an!S!p qf $    ? F FVTaMbcdMeNTUbNcdeNf!G!h i    ; B BQ^ B _ ` a))L)Kxx)3/55a8H 66-44hT_4`  %osD%cPttjjtd5}t j |}dddd}tj||jd}tj|d|}t||df}|S#1swYYxYw)z Harmonize the SQL statement for genotype table only queries to include columns to UNION with genotype and allele table queries. JOIN genotype table to allele table on locus_id to include ref columns values. zreturn_columns_genotype.jsonNz6ON\s+`(\w+)`\.`(a_id)`\s+=\s+`(\w+)`\.`(a_id)`\s+WHEREr1z*ON `\1`.`locus_id` = `\3`.`locus_id` WHEREa_id)r) openospathrVrjsonloadrAsearchrDrWr])rHinfilegenotype_return_columnsjoin_condition_regex allele_tables rharmonize_germline_sqlrjys bggll13QR S4W]"&))F"34U99137==a@L &&%'UWZ [C 16MT`bhSi jC J44s BB%cpg}|D].}i}|D]}||vr ||||<d||<|j|0|S)zy Harmonizes raw query results to include all columns in fields_list. Columns not in fields_list have value None. N)r)results fields_listharmonized_resultsresultharmonized_resultfields rharmonize_germline_resultsrrsh5  0E+1%=!%(+/!%(  0 !!"345 rc td|D}g}|sy|D]!\}}}|jd|g|||dgd#|dddd d d id d igdd|ddd|dddd|ididS)zu Create a payload to query locus_id/ref pairs from the allele table for genotypes missing ref column values. c3DK|]}|d |d|d|dfyw)rNr0 chromosomestarting_positionr ).0rs r z+get_germline_ref_payload..s2rQRabchaiaqQz]AlOQ7J5KLrs  Ninchrstartend conditionvalues geno_binsproject_contextFTandr0allele$locus_idr allele$ref assay_filters raw_filtersidname allele$a_idrrr)radjust_geno_binsrRlogicfields is_cohortr)setr)rlgenotype_payload locus_idsallele_filtersr0r|r;s rget_germline_ref_payloadrsrV]rrIN ' #sj"%C@A   ,,=>! * + L !  &}5oFtL(7HP!>  rcp|dDcic] }|d|d }}|D]}|d ||d|d<ycc}w)zW Update genotype results with ref column values from locus_id/ref query result rlr0rNr )rl locus_id_refsrolocus_id_ref_maps rupdate_genotype_only_refrscIVV_H`afz*F5M9aa= %= $ ( );<u =bs3c|ddddddiddid d id d igdd |dd d|dd dddg|Dcgc]}|d|d |d dc}dgididScc}w)zZ Create a payload to query locus ids from the allele table with a location filter rFTrr0rruz allele$chrrvz allele$posrrrrrrrrzr{rrrr ) locationsrlocations rget_germline_loci_payloadrs ,,=>! * + < ( , / L !   &}5oFtL(7HP!%)"$ /8 & #+(0 '=)12E)F'/0C'D& % $   (&sAc i}|D](}td|d|d|d|dddd||d<*|D]}||ddj|d !|S) z] Produces a dictionary with locus_id as key and a set of samples and entry as value. Nr0rurvr)r.r0rurvrr<)samplesentryrr3)radd)lociresults_entries loci_dictlocusrs r_produce_loci_dictrsI  u!!*-#L1%*+>%?U|  ( % #$  !H% #$Y/33E+4FGH rct||}g}|D]3}|D],}|||dvs|jd|i||dd|i.5||zS)a If the result_entries does not contain entry with sample_id of specifific starting_position the the genotype type is either no-call or ref. Args: samples: list of all samples loci: list of information on each loci within the filter e.g. { "locus_id": "1_1076145_A_T", "chromosome": "1", "starting_position": 1076145, "ref": "A", } result_entries: list of results from extract_assay query. e.g. { "sample_id": "SAMPLE_2", "allele_id": "1_1076145_A_AT", "locus_id": "1_1076145_A_T", "chromosome": "1", "starting_position": 1076145, "ref": "A", "alt": "AT", "genotype_type": "het-alt", } type_to_infer: type to infer either "ref" or "no-call" Returns: list of infered entries with added inferred genotype type and other entries retrieved from result for loci of interest. rr3rr )rr)rrresult_entries type_to_inferrinferred_entriesrsamples rinfer_genotype_typer s8#48I  FYu-i88 ''#V#E*73(   , ,,rc<|Dcgc] }|||vs |c}Scc}w)a3 Filters results by key and restricted_values. Args: results: list of results from extract_assay query. e.g. { "sample_id": "SAMPLE_2", "allele_id": "1_1076145_A_AT", "locus_id": "1_1076145_A_T", "chromosome": "1", "starting_position": 1076145, "ref": "A", "alt": "AT", "genotype_type": "het-alt", } key: key to filter by restricted_values: list of values to filter by Returns: list of filtered entries r )rlkeyrestricted_valuesrs rfilter_resultsr4s$* ' Ne%*rs" 5, ;Dk& ! !  !- 9^<&  !H=#L0(- (-#(-5?(-PS(-(-VO O!O6:OOr