Hydroxymethylation analysis of cell-free nucleic acid samples for assigning tissue of origin, and related methods of use

Abstract

A method is provided for probabilistically assigning a tissue of origin to a nucleic acid in a sample, e.g., DNA in a cell-free fluid sample obtained from a human subject. A hydroxymethylation profile is generated for the sample DNA and then compared across a reference data set of hydroxymethylation profile vectors, where each hydroxymethylation profile vector identifies the hydroxymethylation profile at a specific reference locus, the tissue-specific gene associated with the reference locus, and the tissue with which the gene and reference locus are associated. A tissue of origin can be probabilistically assigned to the sample nucleic acid using the results of the comparison. Other methods of use are also provided.

1 . A method for probabilistically assigning a tissue of origin to a nucleic acid in a cell-free fluid sample obtained from a human subject, comprising:
 (a) generating a hydroxymethylation profile of the nucleic acid;   (b) comparing the hydroxymethylation profile of the nucleic acid with each of a plurality of reference hydroxymethylation profiles in a reference data set, each reference hydroxymethylation profile corresponding to a reference locus contained within a genomic region associated with a tissue-specific gene;   (c) identifying at least one reference locus having a reference hydroxymethylation profile that is substantially similar to the detected hydroxymethylation profile in (a); and   (d) assigning at least one tissue of origin to the nucleic acid based on the at least one reference locus identified in (c) and the corresponding tissue-specific gene.   
     
     
         2 . The method of  claim 1 , wherein each reference locus is contained within a differentially hydroxymethylated genomic region associated with a tissue-specific gene. 
     
     
         3 . The method of  claim 2 , wherein the reference set comprises hydroxymethylation profiles of reference loci associated with a plurality of tissue-specific genes. 
     
     
         4 . The method of  claim 3 , further comprising, between (c) and (d), (c′) determining from the substantial similarity a probability that the nucleic acid has a tissue of origin corresponding to the at least one reference locus. 
     
     
         5 . The method of  claim 4 , wherein the substantial similarity is detected from a hydroxymethylation difference data set that comprises differences identified between the generated hydroxymethylation profile of the nucleic acid and each of the plurality of reference hydroxymethylation profiles. 
     
     
         6 . The method of  claim 1 , wherein the differentially hydroxymethylated genomic region is functionally associated with the corresponding tissue-specific gene. 
     
     
         7 . The method of  claim 6 , wherein the reference set comprises a hydroxymethylation vector that identifies at least the following: the hydroxymethylation profile at a specific reference locus; the tissue-specific gene associated with the reference locus; and the tissue with which the gene and reference locus are associated. 
     
     
         8 . The method of  claim 7 , wherein the hydroxymethylation profile at each reference locus includes at least one of:
 hydroxymethylation density within the reference locus;   total 5-hydroxymethylcytosine residues within the reference locus;   location of the 5-hydroxymethylcytosine residues within the reference locus;   identification of a hydroxymethylated site as hemi-hydroxymethylated or fully hydroxymethylated; and   relative positions of 5-hydroxymethylcytosine residues within the reference locus.   
     
     
         9 . The method of  claim 8 , wherein the hydroxymethylation profile at each reference locus includes hydroxymethylation density. 
     
     
         10 . The method of  claim 9 , wherein the hydroxymethylation density comprises the ratio of 5-hydroxymethylcytosine residues to the total modified cytosine residues and unmodified cytosine residues within the reference locus. 
     
     
         11 . The method of  claim 9 , wherein the hydroxymethylation density comprises the fraction of cytosines at CpG sites that is hydroxymethylated. 
     
     
         12 . The method of  claim 8 , wherein the at least one reference locus is contained within a gene body or a component thereof. 
     
     
         13 . The method of  claim 12 , wherein the at least one reference locus is contained within an intron or exon. 
     
     
         14 . The method of  claim 8 , wherein the at least one reference locus is contained within a genomic annotation feature outside of the gene body. 
     
     
         15 . The method of  claim 14 , wherein the at least one reference locus is contained within a promoter, an enhancer, a transcription initiation site, a transcription stop site, a DNA binding site, or a combination thereof. 
     
     
         16 . The method of  claim 12 , wherein the at least one reference locus comprises a DNA binding site. 
     
     
         17 . The method of  claim 16 , wherein the DNA binding site comprises a silenced region. 
     
     
         18 . The method of  claim 16 , wherein the DNA binding site comprises a transcription factor binding site. 
     
     
         19 . The method of  claim 18 , wherein the transcription factor binding site comprises a transcription repressor binding site. 
     
     
         20 . The method of  claim 19 , wherein the DNA binding site comprises a CTCF binding site. 
     
     
         21 . The method of  claim 1 , further including determining at least one of: nucleic acid length; nucleic acid fragment distribution; methylation profile; and nucleosome positioning. 
     
     
         22 . The method of  claim 1 , wherein the plurality of reference hydroxymethylation profiles in the reference set comprise a hydroxymethylation profile of at least one reference locus associated with at least one gene originating from a human tissue type comprising: adipose; adrenal gland; bladder; bone marrow; brain; breast; colon; cerebral cortex; cervical; uterine; digestive; endometrial; epididymal; esophageal; Fallopian tube; gall bladder; gastrointestinal; heart muscle; hypothalamus; kidney; liver; lung; lymph nodes; ovary; pancreas; parathyroid gland; placenta; prostate; salivary; seminal vesicle; skeletal muscle; smooth muscle; skin; spleen;
 stomach; testis; thyroid gland; tonsil; or a combinations thereof.   
     
     
         23 . The method of  claim 22 , wherein the plurality of reference hydroxymethylation profiles in the reference set comprise a hydroxymethylation profile of at least one reference locus associated with at least one gene originating from a human tissue type comprising: bladder; bone marrow; brain; breast; colon; gastrointestinal; heart muscle; hypothalamus; kidney; liver; lung; lymph nodes; ovary; pancreas; placenta; prostate; skin; smooth muscle; testis; or a combination thereof. 
     
     
         24 . The method of  claim 22 , wherein the plurality of reference hydroxymethylation profiles in the reference set comprises, for each of a plurality of human tissue types, at least one reference locus associated with each of a plurality of tissue-specific genes. 
     
     
         25 . The method of  claim 7 , wherein the reference set comprises hydroxymethylation vectors for at least three tissue types. 
     
     
         26 . The method of  claim 25 , wherein the reference set comprises hydroxymethylation vectors for at least five tissue types. 
     
     
         27 . The method of  claim 26 , wherein the reference set comprises hydroxymethylation vectors for at least 10 tissue types. 
     
     
         28 . The method of  claim 27 , wherein the reference set comprises hydroxymethylation vectors for at least 30 tissue types. 
     
     
         29 . The method of  claim 24 , wherein the reference set comprises hydroxymethylation vectors for at least 5 genes per tissue type. 
     
     
         30 . The method of  claim 29 , wherein the reference set comprises hydroxymethylation vectors for at least 20 genes per tissue type. 
     
     
         31 . The method of  claim 1 , wherein each reference locus is contained within a differentially represented gene comprising AARD, ADARB1, AKR1B10, ANAPC7, APCDD1L-AS1, APOA1BP, CALHM2, CASC3, CLEC2L, COL4A5, CRYM-AS1, EPHA3, FAHD2B, FKBP9L, FOPNL, GNG11, GSN, GSTTP2, GTSF1, IAH1, KCNMB1, KIAA1644, LAMC1, LINC00310, LOC100506394, LOC100507066, LOC493754, MAP1B, MGC27382, MIR5695, NENF, NT5DC3, P2RX1, PCP4, PGM5, PLCD4, PTGFR, RBFOX3, RPL30, SCARA3, SLITS, SNX29P1, SPATA4, ST8SIA1, TBX4, TXNRD1, VCL, VPS72, WFDC3, ZNF791, ADAMTS20, AFF2, ANKRD18B, ANKRD18DP, ANKRD20A5P, ANKRD20A8P, ANKRD26P1, ANKRD3OBP2, ANKRD34B, ANKRD34C, ATP6V0B, AVPR1A, BLOC1S3, BOLL, BRIP1, C5orf63, CA1, CALB1, CALCR, CCDC39, CCDC96, CDCA5, CDH12, CDH9, CEACAM8, CENPE, CENPK, CFL1, CHAT, COMMD5, COMMD8, CRISP1, CRISP2, CSN1S1, CSN1S2AP, CT64, CXCR2, CXorf30, DAZL, DCDC1, DRGX, DSG1, DUT, DUX2, DUX4L2, DUX4L3, DYTN, EDA2R, EFCAB3, EPOR, EVX2, F13A1, FCAR, FCER1A, FKBP1B, FOXA2, GABPB1-AS1, GABRA1, GABRA4, GABRG1, GALNTL6, GFER, GFI1, GPR152, GPR158-AS1, GPRC6A, GRM1, GRP, GRXCR1, GYPA, HAGHL, HCRTR2, HEPHL1, HJURP, HK3, HMBS, HORMAD2, ID4, IL17RA, IL18RAP, KCNC2, KCNK18, KIF18A, KIF20B, KLHL1, LEF1-AS1, LHX1, LHX3, LHX8, LINC00273, LINC00558, LINC00645, LINC00648, LLPH, LOC100129620, LOC100270679, LOC100505776, LOC100506422, LOC284801, LOC392232, LOC399815, LOC401074, LOC401134, LOC440040, LOC440970, LOC642236, LOC727924, LOC728012, LPO, LRRIQ1, MAGEA11, MAGOHB, MEFV, MIF4GD, MIR548A3, MIR5692A1, MMP8, MMP9, MMRN1, MRGPRE, MS4A3, MSRB1, MYO3A, NCOR1P1, NDST3, NDST4, NHLRC4, NOXRED1, NPAP1, NUF2, OR8G1, OR8U8, ORC6, OSM, OTX2, PAX3, PBK, PCDH8, PDCL2, POTEA, PROK2, PRR7, RAB24, RAX, RBM46, RGS18, RGS19, RIMS2, RNF175, RRM2, SATB2-AS1, SCN11A, SCRT2, SERPINB10, SGOL2, SIX1, SKA3, SKOR1, SLC22A16, SLC4A10, SLC5A7, SLC5A8, SLCO6A1, SNAP91, SPAG16, SPATA16, SPDL1, SPINK13, STPG2, STXBPSL, TARM1, TDRDS, TEX15, THSD7B, TMPRSS11A, TMPRSS11B, TMPRSS11D, TRIM58, TUBB4B, UNC13C, USP6, VRK1, VSTM1, VWDE, WDR49, WFDC8, WFDC9, ZICS, ZNF230, ZNF300P1, ZNF470, ZNF502, ZNF599, ZPBP, C1orf159, CCDC42B, CCDC8, CD248, CERCAM, CNTN2, CRB2, EVISL, FZD9, HAPLN2, HPDL, LINC00575, LOC284950, LOC339666, LRIT1, PLEKHH1, SHISA2, SLC46A2, TFAP2E, TMEM235, TNFRSF18, ZFP36, C1orf109, CACNA1F, COX7B, EFNB1, FGL2, F1134208, GNRH1, GSTT1, IL12A, KCND1, KDELR1, LAT, LOC100130992, LOC100287042, LOC401242, MRPL10, NT5C3B, PDZD4, PPYR1, RAD51D, RBMXL3, RENBP, SCNN1B, SERPINB5, SLC9A6, XBP1, ZNF189, ABR, ADPRHL1, ASB18, ATP1A3, CCDC74A, CDH13, CHRNA10, CORO6, FSD2, GALNT16, GDAP1L1, GJA3, GLUD1P3, GPRC5B, LOC100506343, LRRC37A6P, LRRC4, MUL1, MYOM2, NDUFB8, NT5DC2, PCDHGA1, PCDHGA10, PCDHGA11, PCDHGA12, PCDHGA2, PCDHGA3, PCDHGA4, PCDHGA5, PCDHGA6, PCDHGA7, PCDHGA8, PCDHGA9, PCDHGB1, PCDHGB2, PCDHGB3, PCDHGB4, PCDHGB5, PCDHGB6, PCDHGB7, PXDNL, RHPN1-AS1, RPL3L, SGCG, TIMM21, TNNT1, TOR4A, UBAC1, WDR37, ZNF12, ZNF213, ZNF550, ZNF843, ZNF844, AGAP6, ASTN1, BBS2, C18orf56, C19orf45, CINP, DGKI, DPY30, DROSHA, FAM169A, FAM66D, FONG, GALNT15, GDAP1, GHRH, GUSBP4, HIF1AN, IFIT1B, INTS2, JAKMIP2-AS1, KGFLP2, LINC00297, LINC00707, LOC100288069, LOC100507140, LPHN3, LYPD1, LZIC, MIR548T, MRAP, NTRK3, PDE6D, PPP1R17, PRSS54, PRSS55, PTPRA, RGS7, SNRNP35, SUGT1P1, UBE2Z, WDR47, WDYHV1, ZNF114, ZNF556, ZNF610, ABP1, AGPAT9, APH1B, ASAP2, ATP4B, ATP6V1A, BBOX1, BCRP3, C11orf63, C13orf35, C16orf96, C19orf21, CA12, CCDC64, COL4A3, CXCL14, CYS1, EIF4E2, EPB41L1, EVC, FIGNL2, FUT3, GALNT14, GLIS2, GUSBP11, HAVCR1, HOXCS, INADL, KATNA1, L2HGDH, LOC100130238, LOC100506305, LOC284100, LOC654433, LYG1, MORN4, MTNR1A, PAQRS, PARG, PAX2, PAX8, PLA2G15, POU5F1, PRKAB1, RNF113B, SENP8, SETD3, SLC22A2, SLC25A30, SLC9A3, SULT1C2, TBC1D7, TSPAN33, ABCB4, ABCC11, ABCC2, ABCGS, ABCG8, ACADSB, ACSMS, AGMO, AGXT2L1, AKR1C2, AKR1C4, AKR1D1, AMDHD1, APOB, APOH, APOM, AQP9, ARHGEF40, ASGR2, ASPG, ATP2B2, BCO2, BDH1, C3, C3P1, C5, CBA, C8B, C8orf74, CASA, CABP2, CALR3, CD7, CDK10, CEACAM16, CES1, CFH, CFHR4, CHP1, CIDEB, CLPTM1L, CLRN3, CPN2, CREB3L3, CROCCP2, CYP2B6, CYP2C18, CYP2C19, CYP2C9, CYP2E1, CYP3A5, CYP7A1, DGAT2, DHODH, DPYS, DRG2, ECHS1, EDC4, EI24, ENO1, F13B, F2, F7, F9, FASN, FETUB, GCH1, GCKR, GDPD4, GFRA1, GHR, GLT1D1, GPAM, GSDMB, HAL, HAO1, HAPLN4, HPD, HPX, HYLS1, IGF2, IGF2-AS, IGFBP1, ISY1-RAB43, ITIH1, ITIH2, ITIH4, IVD, KLC4, LBP, LEFTY1, LIPG, LOC100288122, LOC284865, LPAL2, LRRC16B, MASP1, MGMT, MST1, MTTP, NAT9, NBR2, NELFE, NR1I2, NUP88, OSGIN1, PAH, PARD6A, PCSK9, PEX19, PGLYRP2, PHGDH, PHYH, PKLR, PLA2G12B, PLG, PNPLA3, POFUT1, POLR1E, PON1, PPL, PRAP1, PRKAG3, PROC, PUS3, RANBP10, RCE1, RND1, RNF123, RORC, RPUSD4, SAA2-SAA4, SCP2, SEC16B, SERPINA6, SERPINC1, SKIV2L, SLC13A5, SLC22A25, SLC25A20, SLC27A5, SLC30A10, SLCO1B3, SMLR1, SNAPC5, SPP2, SRD5A1, STAT2, STEAP3, SULT2A1, TFR2, THNSL2, TIAF1, TM6SF2, TMEM45A, TMPRSS6, TTC31, TTC38, UROC1, XYLB, ZCCHC9, ZSCAN22, ADAMS, ANKRD50, ARGLU1, ARL6, ARSJ, BMP5, BMPR2, BTG3, C1orf140, CALM2, CCDC102B, CCNL1, CCR5, CD36, CHN1, CLIC2, CPEB2, CRBN, CYP4Z2P, CYYR1, DACH1, DGKE, DGKH, DISP1, DOCK4, ETV1, EXOC1, FAM204A, FAT4, FGD5-AS1, F1134503, FRYL, GBP1P1, GNB4, GPR110, GPR116, HMCN1, HMGN1, IFI44, IL15, ITGA2, KAL1, KDR, KITLG, KLHL41, LDB2, LINC00032, LINC00240, LINC00551, LINC00657, LOC100131234, LOC100505495, LOC100507217, LOC643733, LPAR6, MGP, ODF2L, PEAK1, PKIA, PLEKHA1, PLEKHG7, PTPRB, QKI, RAD21, RALA, RAP2A, RCC1, SAMD12, SESTD1, SH3GLB1, SKAP2, SLC35A5, SMURF2, SPRED1, SRSF1, TCF4, TIGD4, TMEM207, TMOD3, UHMK1, VEGFC, XIST, YIPF5, ZC2HC1A, ZEB1, AKNA, ANKRD34A, C14orf183, CCDC107, CD180, CD3G, CD74, CDC42SE2, CHMP7, COTL1, CYTH1, FAIM3, FAM65B, GPX4, GSTP1, HLA-DMA, HLA-DOA, HLA-DPB1, HVCN1, ICAM2, ICOS, IL6, ITGB7, LOC100130557, MDM4, METTL21D, MGC16275, MIR548AN, NAPSB, RPL39L, RPS11, SEPT6, SH2D3C, TAP1, TEAD2, TMEM60, TNFRSF9, TRAF1, UBAC2, UCP2, WDR87, ACTA2, ADAMTSS, AQP11, ATP1B2, BICD1, C7, CDH3, CDON, COL14A1, HS3ST1, KLF4, LEMD1-AS1, LINC00672, LOC100129617, LOC339298, LRRC17, NDP, NTF3, OMD, PDGFRA, PGR, ROBO4, RWDD4, SCDS, SERPINE2, SLC25A17, SNCAIP, SYTL4, TENM4, TSPANS, UBXN8, ZNF93, ALDH1L2, ANKEF1, ASUN, B3GALNT2, BBIP1, C6orf201, CASP9, CCDC110, CCDC65, CDK14, CELA2B, CELA3B, CFTR, CHMP4C, CHRNA7, CLCN3, CLDN1, CPA3, CUZD1, DNAJC10, DNAJC3, EIF2S1, EIF4EBP1, ERO1LB, FAIM, FAM160A1, FAM162A, FAM221A, FAM24B-CUZD1, GARS, GUCA1C, HSPA13, IFRD1, INTS6, KCTD16, LINC00339, LMAN1, LOC154092, LOC201651, LOC644838, LYPD6B, MAN1A2, MGC72080, MKNK1, MPP6, MSRB2, NAA16, NOMO2, OSTC, PEX7, PGRMC2, PIK3CB, PLA2G12A, PM20D1, PPP1R9A, PRRC1, PRSS3, RNPC3, SCFD1, SCRN3, SERPINI2, SH3YL1, SLC16A7, SLC33A1, SLC4A4, SORBS2, SRBD1, SSR1, TDH, TDP2, TMEM51-AS1, TMEM65, TRHDE, TRIM44, UGT2A3, UMOD, WDFY2, WHAMM, XPOT, ACER2, ADAM12, ADCY10, ADCY7, ADORA2B, AIM1L, AKNAD1, ALDH3B2, AMOT, ANGPT2, APOLD1, ATF3, ATG9B, ATP6V1C2, ATRIP, BCAR3, BCAR4, BEAN1, BPGM, BTBD19, C11orf70, C16orf46, C16orf74, C17orf98, C1QTNF6, C2orf62, C2orf83, C3orf52, C4orf26, C4orf51, C6orf99, C7orf71, C9orf129, CACNA2D3-AS1, CAPN6, CCDC125, CEP41, CLEC7A, CPA4, CSF3R, CTSL3P, CXorf56, CYP19A1, DACT2, DDX59, DLX3, DLX5, DLX6-AS1, DNMT1, DUSP4, EBI3, EDARADD, EPAS1, EPS8L1, ERVFRD-1, ERVMER34-1, ETV3, ETV4, EXTL1, EZR, FAM184A, FAM89A, FANCE, FBLN1, FBN2, FHDC1, FOLR1, GATA2, GATA3, GBA, GCM1, GDPD3, GLDN, GM2A, GPR156, GSG1, GSTA3, GSTA4, HELLS, HOPX, HSD11B2, HSPBAP1, IGF2BP3, IGSF5, ISM2, KATNBL1, KIAA1467, KIAA1609, KISS1, KLRG2, L1TD1, LCMT1, LCTL, LIN28B, LINC00439, LOC100131564, LOC100506746, LOC100527964, LOC151475, LOC152578, LOC284551, LOC643441, MB21D2, MED12, METTL21C, MFSD2B, MGC16121, MINA, MORC4, MSANTD3, MUTYH, NOS3, NVL, NXF1, OLR1, OSCP1, OSTCP1, P2RY6, PCBP1-AS1, PDE6A, PGC, PGF, PLA1A, PLAC4, PPP1R14D, PSG11, PSG2, PSG6, PSG7, PSG8, PTGES, PVRL3-AS1, PVRL4, PWWP2B, RAB36, RHO, RNF222, RPSAP58, RS1, SBF2-AS1, SCIN, SDC1, SEMA3F, SEMA6D, SEPT12, SH2D7, SH3GLB2, SLC13A4, SLC26A2, SLC2A1-AS1, SMAGP, SMARCB1, SMIM13, SNX12, SP6, SPESP1, SPIRE2, ST3GAL6-AS1, SVEP1, SYT8, TENM3, TLR3, TMEM216, TMEM218, TMEM52B, TMPRSS7, TP63, TPRXL, TRIM29, TRPVS, TUFT1, USHBP1, VAMPS, VGLL1, WBP2NL, XRCC2, ZDHHC1, ZNF320, ZNF331, ZNF354B, ZNF468, ZNF525, ZNF702P, ZNF90, ZP3, ZSWIM2, ZSWIM7, ABCC4, ACPP, ALDH1A2, ANKRD66, AP1B1P1, ARG2, BEND4, C1orf85, C6orf132, C6orf52, CHRNA2, COQ7, EVX1, EYA2, F1139080, GDEP, HMGN2P46, ISX, KLKP1, MCCC2, MEAF6, MFSD4, MRPS23, NCAPD3, OACYLP, OAZ3, OR51E2, PAK1IP1, PCAT1, PI15, PPP1R7, PSMA4, SCNN1G, SLC14A1, SLC26A3, SLC2A12, SLC30A4, SLC35F2, TMEM79, TPM3P9, TTC12, USP50, ZNF350, ZNF532, ZNF589, AKIP1, ARHGEF26-AS1, ATG4A, BOLAS, CHMP4A, DHH, EPPK1, FCHSD1, GADL1, GCC1, GIPC3, HIGD2B, HTR3C, LINC00654, LOC91948, OLFML3, PTGES3L, WSCD2, ZNF202, ZNF205-AS1, ZNF426, ACOT1, C12orf54, CREB3, DDN, FAM50A, FNDC4, GORASP1, GUCA1B, IMPDH2, LINC00284, LOC100128076, LOC100128682, LOC100131434, LRP10, LRRC55, LRRC73, NOTO, RNF112, RTBDN, SIGLEC8, TNN, TRIM50, UBE2T, ZNF391, ZNF792, ALX4, AQP2, ATAD3B, C11orf85, CABLES2, CCDC114, CCDC27, CCNF, CD300C, CILP, DES, DMRT1, DNAJB8-AS1, DRD4, EFCAB6-AS1, FAM153B, FAM163B, FGD2, HSD17B3, IFI27L1, IGFN1, ITPR3, KCNK15, KIFC2, KRT78, KRT82, LOC100506385, LOC283914, LOC284661, LOC399829, LOC653712, LOC728716, MAGED2, MYH16, MYLK2, NPTX2, PLK5, PMS2CL, PNPLA1, PODNL1, POLD1, PRMT8, PTPRVP, RD3, RIMS4, RNU6-81, RPL26L1, SALL4, SERPINA10, SPATA31E1, STAB1, SULT4A1, TCERG1L, TEKT4P2, THEG, TMEM130, TPO, UPK3B, WASH3P, WRAP53, ZNF280B, or ZNF793. 
     
     
         32 . The method of  claim 1 , wherein each reference locus is contained within a differentially represented, differentially expressed gene comprising AARD, ADARB1, AKR1B10, CRYM-AS1, EPHA3, GTSF1, KCNMB1, MAP1B, NT5DC3, P2RX1, PCP4, PGM5, PLCD4, PTGFR, RBFOX3, SCARA3, SLITS, SNX29P1, ST8SIA1, TBX4, TXNRD1, VCL, WFDC3, ABCA13, AZU1, CA1, CEACAM8, CLEC6A, DAZL, DRGX, DYTN, ELANE, FCAR, GFI1, GYPA, HK3, IL18RAP, LINC00333, LINC00550, LINC00558, LIPN, LOC100129620, LOC442028, LPO, MEFV, MMP8, MMP9, MPO, MS4A3, MYB, MYO1F, NCOR1P1, NME8, OR8U8, OSCAR, PARPBP, PAX3, PDCL2, PRDM13, PROK2, RGS18, RGS21, RRM2, RXFP2, SERPINB10, SLC22A16, SPATA16, SPI1, SPTA1, TARM1, TMPRSS15, TRIM58, VSTM1, XKR3, ASPHD1, BRSK1, CCDC177, CNTN2, HAPLN2, KCNJ10, KIF1A, LGI3, NCAN, NR2E1, RHBDL3, TMEM151A, TMEM235, TMEM59L, ZNF488, FGL2, SCNN1B, ADPRHL1, ASB18, ATP1A3, CDH13, CORO6, FSD2, GALNT16, GJA3, MYOM2, PCDHGA1, PCDHGA10, PCDHGA11, PCDHGA12, PCDHGA2, PCDHGA3, PCDHGA4, PCDHGA5, PCDHGA6, PCDHGA7, PCDHGA8, PCDHGA9, PCDHGB1, PCDHGB2, PCDHGB3, PCDHGB5, PCDHGB6, PCDHGB7, PXDNL, RPL3L, SGCG, TNNT1, ASTN1, DGK1, BBOX1, BCRP3, CA12, COL4A3, CXCL14, CYS1, EVC, FIGNL2, GALNT14, GLIS2, HAVCR1, LOC100130238, PAQR5, PAX2, PAX8, PLA2G15, SLC22A2, SLC9A3, SULT1C2, TSPAN33, ABCC2, ABCG5, ABCG8, AHSG, AKR1C4, AMDHD1, APOH, ASGR1, ASGR2, C3P1, CBA, C8B, C9, CASA, CFHR4, CPN2, CREB3L3, CYP2B6, CYP2C9, CYP2E1, CYP3A4, DHODH, F12, F2, F7, GC, HAL, HPD, HPX, INHBC, ITIH1, ITIH2, ITIH3, KNG1, LBP, PGLYRP2, PLG, PROC, PRODH2, SERPINA6, SERPINC1, SLC13A5, SLC22A1, SLC25A47, SLC27A5, SULT2A1, TFR2, TMPRSS6, UROC1, ANXA3, BMP5, CYP4Z2P, DRAM1, FMO2, GNA15, IDOL KCNS3, LIMCH1, MBIP, PAPSS2, PCDH17, RSPH4A, RTKN2, TRPC6, ACAP1, AKNA, ARHGAP9, BCL11B, BIRC3, CARD11, CD180, CD37, CD3E, CD3G, CD40, CETP, CIITA, CLEC2D, CNR2, CXCR5, DOCK10, ETS1, GPR132, GPR18, ICAM2, ICOS, ITGB7, ITK, KIAA1551, KLHL6, LCK, LINC00426, LY86, MDS2, MEOX2, MX2, NAPSB, PATL2, PCED1B-AS1, PIK3CD, PLVAP, POU2F2, SCIMP, SCML4, SLFN12L, SMAP2, SP110, SPIB, TLR10, TMC8, TMEM156, TNFAIP8, TNFRSF9, TNFSF11, TRAF1, TRIM22, UCP2, ZC3H12D, ABCA10, ACSS3, ADAMTS5, AQP11, C7, CDH3, CDON, CLDN11, COL14A1, COLEC11, ESR2, FAM198B, FZD3, GALNT10, GLI2, GSTM5, HS3ST1, LEMD1-AS1, LHX9, LRRC17, MCHR1, MRC2, NTF3, OMD, PDGFRA, PGR, PKNOX2, PTCH2, RWDD4, SCD5, SERPINE2, SIMC1, SLC25A17, SNCAIP, SULF2, TENM4, TSPAN5, TTC8, UBXN8, ALDH1L2, ANKEF1, ASNS, B3GALNT2, BCAT1, CASP9, CCDC110, CELA2B, CELA3B, CFTR, CHMP4C, CLDN1, CNIH3, COCH, CUZD1, DEFB1, EGF, EIF4EBP1, ERP27, FAM24B-CUZD1, FBXW12, GUCA1C, KCTD16, KIAA1324, LINC00339, LNX2, MKNK1, NAA16, NPHS1, PAIP2B, PM20D1, PRSS3, SCGN, SEL1L, SERPINI2, SH3YL1, SLC33A1, SLC4A4, TC2N, TDH, TMEM51-AS1, TRHDE, UMOD, ADAM12, ATG9B, ATP6V1C2, BCAR4, BMP1, BPGM, CSF3R, CYP11A1, CYP19A1, DACT2, DEPDC1B, DLX5, EBI3, GCM1, GPR78, GSTA3, IGF2BP3, IGSF5, ISM2, KISS1, KRT23, LIN28B, MMP11, PGF, PSG11, PSG2, PSG3, PSG6, PSG7, PSG8, SPTLC3, TPRXL, ZFAT, ZNF554, ABCC4, ACPP, ALOX15B, ANO7, AP1B1P1, ARG2, BEND4, C6orf132, C6orf52, CANT1, CASZ1, CBFA2T2, CHRM1, CHRNA2, COL26A1, EVX1, EYA2, FAM135A, HMGN2P46, KLK2, KLKP1, MME, MPPED2, MUC12, NANS, NCAPD3, NEFH, NIPAL3, OACYLP, OR51E2, PAK1IP1, PDE9A, PI15, PMEPA1, POTEF, RDH11, SCNN1G, SIM2, SLC14A1, SLC22A3, SLC2A12, SLC30A4, SLC35F2, SLC37A1, SLC39A6, SPDEF, STEAP2, THSD4, TMEM79, TPM3P9, TXNDC16, URB1, VIPR1, ZNF350, ZNF532, ZNF613, ZNF649, ZNF761, ZNF827, GADL1, GIPC3, PTGES3L, WSCD2, CCDC27, DMRT1, DNAJC5G, FBXO24, IGLL1, MOV10L1, SEPT14, THEG, or TTC16. 
     
     
         33 . The method of  claim 1 , further comprising determining the tissue of origin of multiple DNA fragments in the cell-free fluid sample. 
     
     
         34 . A method for detecting a condition of a tissue or organ in a human subject, comprising:
 (a) generating a hydroxymethylation profile of DNA in a cell-free fluid sample obtained from the subject;   (b) comparing the hydroxymethylation profile generated in (a) with each of a plurality of reference hydroxymethylation profiles in a reference set, each reference hydroxymethylation profile associated with (i) a reference locus associated with a tissue-specific gene and (ii) a condition of a specific tissue or organ; and   (c) based on the hydroxymethylation profile generated for the subject relative to the reference hydroxymethylation profiles, identifying the condition of the specific tissue or organ in the subject.   
     
     
         35 . The method of  claim 34 , wherein the reference set comprises at least one reference locus on each of a plurality of tissue-specific genes. 
     
     
         36 . The method of  claim 35 , further comprising making a diagnosis, a treatment decision, or a prognosis based on the detected condition. 
     
     
         37 . A method for detecting a change in the condition of a tissue or organ in a human subject, comprising:
 (a) generating an initial hydroxymethylation profile at one or more loci of DNA fragments in a cell-free fluid sample obtained from the subject, where the hydroxymethylation profile is associated with an initial condition of a tissue or organ;   (b) at a later time, generating a subsequent hydroxymethylation profile at the same one or more loci in a cell-free fluid sample obtained from the subject; and   (c) comparing the subsequent hydroxymethylation profile with the initial hydroxymethylation profile to detect a change in the condition of a tissue or organ.   
     
     
         38 . The method of  claim 37 , comprising repeating step (b) at one or more additional times so as to monitor the progress of a condition, disease, or disorder associated with the tissue or organ. 
     
     
         39 . The method of  claim 38 , comprising assessing the efficacy of a therapeutic intervention. 
     
     
         40 . The method of  claim 38 , comprising monitoring changes in gene expression over time. 
     
     
         41 . The method of  claim 40 , further comprising making a diagnosis, a treatment decision, or a prognosis based on the detected change in condition. 
     
     
         42 . An improved method for analyzing a cell-free DNA sample obtained from human plasma in which cell-free DNA fragments in the sample are isolated, amplified, and sequenced, wherein the improvement comprises identifying contamination of the sample with genomic material originating from peripheral blood mononuclear cells (PBMCs) by:
 (a) generating a hydroxymethylation profile of DNA fragments in the cell-free sample; and   (b) comparing the hydroxymethylation profile generated in (a) with each of a plurality of reference hydroxymethylation profiles in a reference set, each reference hydroxymethylation profile associated with a reference locus corresponding to a tissue-specific gene, wherein the reference hydroxymethylation profiles include at least one PBMC hydroxymethylation profile; and   (c) identifying, from the comparison, the presence of genomic material originating from PBMCs.   
     
     
         43 . The improved method of  claim 42 , further including estimating the level of PBMC genomic contamination from (b) and (c). 
     
     
         44 . The improved method of  claim 43 , further including making a determination as to whether the sample is suitable for cell-free DNA analysis based on the level of PBMC genomic contamination. 
     
     
         45 . The improved method of  claim 42 , wherein the reference set comprises a vector of a plurality of reference hydroxymethylation profiles each corresponding to a gene locus associated with a PBMC sub-type. 
     
     
         46 . The improved method of  claim 45 , further including using the vector to offset PBMC contamination in the cell-free sample. 
     
     
         47 . The improved method of  claim 46 , wherein offsetting PBMC contamination in the cell-free sample comprises using the correction vector to calibrate and normalize at least one of mutation load, fragment count, variant allele frequency, and minor allele frequency. 
     
     
         48 . A method for probabilistically assigning a tissue of origin to cell-free DNA in a sample obtained from a human subject,
 (a) generating a hydroxymethylation profile of DNA in the sample by:
 (i) adding an affinity tag to only 5-hydroxymethylcytosine residues in the cell-free DNA; 
 (ii) enriching for DNA molecules that are tagged with the affinity tag by binding to a support; and 
 (iii) sequencing the enriched DNA molecules to provide a map with each 5-hydroxymethylcytosine site identified; 
   (b) comparing the hydroxymethylation profile generated in (a) with each of a plurality of reference hydroxymethylation profiles in a reference data set, each reference hydroxymethylation profile associated with a reference locus in a differentially hydroxymethylated genomic region associated with a tissue-specific gene,   (c) identifying at least one reference locus in the reference data set having a reference hydroxymethylation profile that is substantially similar to the generated hydroxymethylation profile; and   (d) assigning at least one tissue of origin to each sequenced DNA molecule based on the at least one reference locus identified in (c) and the corresponding tissue-specific gene.   
     
     
         49 . The method of  claim 48 , wherein the reference set comprises at least one reference locus associated with each of a plurality of tissue-specific genes. 
     
     
         50 . The method of  claim 48 , wherein the affinity tag is biotin. 
     
     
         51 . The method of  claim 50 , wherein the method comprises: adding adaptor sequences onto the ends of the DNA; incubating the adaptor-ligated DNA with a DNA β-glucosyltransferase and UDP glucose modified with a chemoselective group, thereby covalently labeling the hydroxymethylated DNA molecules in the DNA with the chemoselective group;
 linking a biotin moiety to the chemoselectively modified DNA via a cycloaddition reaction; 
 enriching for the biotinylated DNA molecules with a biotin-binding support; amplifying the enriched DNA using primers that bind to the adaptors; and sequencing the amplified DNA to produce a plurality of sequence reads. 
 
     
     
         52 . The method of  claim 51 , further prior to (a)(ii), removing the tagged 5-hydroxymethylcytosine-containing DNA from the sample, leaving unmodified DNA and DNA containing unmodified 5-methylcytosine residues. 
     
     
         53 . The method of  claim 52 , further comprising ascertaining the methylation state of the DNA containing unmodified 5-methylcytosine residues. 
     
     
         54 . A method for determining whether a human subject is predisposed to develop an abnormal condition of a tissue or organ, the method comprising:
 (a) generating a hydroxymethylation profile at one or more loci of DNA fragments in a cell-free fluid sample obtained from the subject;   (b) comparing the detected hydroxymethylation profile with a plurality of reference hydroxymethylation profiles in a reference set each associated with (i) a reference locus on a tissue-specific gene and (ii) a condition of the tissue indicating a predisposition to develop an abnormal condition of the specific tissue or organ; and   (c) identifying a predisposition to develop the abnormal condition of the tissue from the comparison in (b).   
     
     
         55 . A method for differentiating nucleic acids associated with different tissues of origin in a sample of cell-free DNA obtained from a human subject, the method comprising:
 (a) generating hydroxymethylation profiles of each of a plurality of nucleic acids in the cell-free DNA sample;   (b) comparing each hydroxymethylation profile with a plurality of reference hydroxymethylation profiles in a reference set each associated with a specific reference locus on a tissue-specific gene; and   (c) differentiating nucleic acids originating from different tissues based on the comparison in (b).   
     
     
         56 . The method of  claim 55 , wherein the reference set comprises two or more reference sets selected from Set 3A, Set 3B, Set 3C, Set 3D, Set 3E, Set 3F, Set 3G, Set 3H, Set 3I, Set 3J, Set 3K, Set 3L, Set 3M, Set 3N, Set 3O, Set 3P, Set 3Q, Set 3R, and Set 3S.