Correlation of nonsilent somatic mutations with the age and the disease stage of 25 people with NKTCL, subjected to whole-exome sequencing.

Transcript

1 Supplementary Figure 1 Correlation of nonsilent somatic mutations with the age and the disease stage of 25 people with NKTCL, subjected to whole-exome sequencing. (a) Distribution of non-silent somatic mutations according to the age of the patients. P value was calculated by two-sided F-test (n = 25 subjects). (b) Distribution of non-silent somatic mutations according to the disease stage of the patients. P value was calculated by twosided Student s t test (mean ± s.d.; n = 25 subjects).

2 Supplementary Figure 2 Distribution of reads mapped to the EBV genome in tumor and matched peripheral blood samples of 25 people with NKTCL, subjected to whole-exome sequencing.

3 Supplementary Figure 3 Somatic copy-number alterations (CNAs) and uniparental disomies (UPDs) in 25 people with NKTCL, subjected to whole-exome sequencing. (a) Types and genomic distribution of somatic copy number alterations (CNAs) and uniparental disomies (UPDs) in each patient. (b) Types and genomic distribution of somatic CNAs and UPDs according to the chromosomes.

4 Supplementary Figure 4 Functional categories of somatic mutations in 25 people with NKTCL, subjected to whole-exome sequencing. (a) Gene ontology analysis of 795 somatic mutations expected to alter the function or structure of the encoded protein. Adjusted P value < 0.01 and False Discover Rate < 0.1 (Online Methods). (b) Functional categories of 26 candidate gene mutations according to 6q21 deletions.

5 Supplementary Figure 5 Relationship between mutations of DDX3X, tumor suppressors, JAK-STAT-pathway molecules and epigenetic modifiers in NKTCL. (a) Distribution of mutations in DDX3X, tumor suppressors (TP53 and MGA), JAK-STAT pathway (STAT3 and STAT5B) and epigenetic modifiers (MLL2, ARID1A, EP300 and ASXL3) in NKTCL. (b) Comparison of the allele frequencies of selected mutations in samples harboring mutations in DDX3X and tumor suppressors. (c) Comparison of the allele frequencies of selected mutations in samples harboring mutations in DDX3X and JAK-STAT pathway molecules. (d) Comparison of the allele frequencies of selected mutations in samples harboring mutations in and DDX3X and epigenetic modifiers. Each axis showed the frequencies of the mutant alleles. When a single gene had 2-Hit mutation, the frequencies of major alleles were indicated. Data were analyzed statistically by two-sided Wilcoxon rank-sum test.

6 Supplementary Figure 6 Cellular localizations of DDX3X and EBV-associated RNA or protein in EBV-infected natural killer cell line KAI3. (a) Cellular localizations of DDX3X and EBV-Encoded RNA (EBER). (b) Cellular localizations of DDX3X and latent membrane protein 1 (LMP-1). Images were visualized by confocal laser scanning microscopy. Scale bar, 5μm.

7 Supplementary Figure 7 Apoptosis assay of DDX3X wild-type natural killer cell line KAI3 transfected with empty vector (vector), wild-type or DDX3X mutants (E348K and A404P). P values were calculated using two-sided Student s t test (mean ± s.d.; n = 3 per group).

8 Supplementary Figure 8 shrna-mediated knockdown of DDX3X enhanced the growth of DDX3X wild-type natural killer cell line KAI3. (a) Flow cytometric plots showing the percentage of GFP + KAI3 cells 3 days (considered as day 0 for cell growth evaluation) and 15 days (considered day 12 for cell growth evaluation) post transduction of cells with scrambled shrna (Scrambled shrna) or shrna targeting DDX3X (DDX3X shrna1). Cells were switched to natural killer culture medium with reduced IL-2 (25 U/mL). (b) Quantification of the percentages of GFP + KAI3 cells transfected with scrambled shrna or DDX3X shrna1 at regular time intervals, as normalized to the levels at day 0. **: P < 0.01 as compared to the scrambled shrna. P values were calculated using two-sided Student s t test (mean ± s.d.; n = 3 per group). (c) Western blot images of DDX3X levels in shrna-transfected, GFP + sorted cells post transduction with the scrambled shrna or DDX3X shrna1. (d) Flow cytometric plots showing the percentage of GFP + KAI3 cells 3 days (considered as day 0 for cell growth evaluation) and 15 days (considered day 12 for cell growth evaluation) post transduction of cells with scrambled shrna (Scrambled shrna) or shrna targeting DDX3X (DDX3X shrna2). Cells were switched to natural killer culture medium with reduced IL-2 (25 U/mL). (e) Quantification of the percentages of GFP + KAI3 cells transfected with scrambled shrna or DDX3X shrna2 at regular time intervals, as normalized to the levels at day 0. **: P < 0.01; ***: P < 0.001, as compared to the scrambled shrna. P values were calculated using two-sided Student s t test (mean ± s.d.; n = 3 per group). (f) Western blot images of DDX3X levels in shrna-transfected, GFP + sorted cells post transduction with the scrambled shrna or DDX3X shrna2.

9 Supplementary Figure 9 Gene-expression profiles of NF- B and MAPK pathways in NKTCL. (a) Significantly upregulated and downregulated genes of NF- B pathway in DDX3X-mutated tumors, as compared to those in DDX3Xwild-type (wt) tumors of NKTCL. (b) Significantly upregulated and downregulated genes of MAPK pathway in DDX3X-mutated tumors, as compared to those in DDX3X-wild-type (wt) tumors of NKTCL.

10 Supplementary Figure 10 Kaplan-Meier analysis of the survival of people with NKTCL, excluding extranasal cases or stage III and stage IV cases. (a) Overall survival (OS) and progression-free survival (PFS) of NKTCL patients excluding extranasal cases according to mutation status of DDX3X and TP53 (upper panel) and according to risk stratification combining International Prognostic Index (IPI) and mutation status of DDX3X and TP53 (lower panel). (b) OS and PFS of NKTCL patients excluding stage III IV cases according to mutation status of DDX3X and TP53 (upper panel) and according to risk stratification combining IPI and mutation status of DDX3X and TP53 (lower panel). Risk stratification combining IPI and mutation status of DDX3X and TP53: Group 1: IPI0 1 and wtddx3x/wttp53; Group 2: IPI0 1 and mutddx3x/muttp53 or IPI2 5 and wtddx3x/wttp53; Group 3: IPI2 5 and mutddx3x/muttp53. wt: wild-type, individuals without DDX3X or TP53 mutations; mut: mutation, individuals with DDX3X or TP53 mutations. One individual with both DDX3X and TP53 mutations was grouped into two cohorts with mutations.

11 Supplementary Figure 11 Variant allele frequency of single-nucleotide variations in 25 people with NKTCL, subjected to whole-exome sequencing. In each plot, the density curve depicted the clustered variant allele frequency (VAF) to determine the number of clusters.

12 Supplementary Figure 12 Flow chart for identification of somatic single-nucleotide variations and indels from paired whole-exome-sequencing data.

13 Supplementary Figure 13 Representative GeneScanning results of TRB or TRG rearrangement in NKTCL. (a) TCRB rearrangement negative. (b) TCRG rearrangement negative. (c) TCRG rearrangement positive. The control gene peaks were shown at 101, 200, 300 and 395nt.

14 Supplementary Figure 14 Representative GeneScanning results of TRD rearrangement in NKTCL. (a) TCRD rearrangement negative. (b) TCRD rearrangement positive. The control gene peaks were shown at 101, 200, 300 and 395nt.

15 Supplementary Figure 15 Positive controls of immunohistochemistry of human anti-relb antibody and human anti-p-erk antibody. (a) The positive control of immunohistochemistry of human colon cancer tissue using anti-relb antibody (Proteintech, AP). (b) The positive control of immunohistochemistry of human lung cancer tissue using anti-p-erk antibody (Cell Signaling, 4370S). Scale bar, 20μm.

16 Supplementary Table 4. Minimal common regions of somatic copy number alterations and uniparental disomies analysis. No. Region Type of alterations Number of alterations Alteration size (Mb) Involved genes 1 1q31.3 UPD NEK7, ATP6V1G3, PTPRC 2 6q21 Deletion q35 Amplification p24.1 Deletion BVES, POPDC3, PREP, PRDM1, ATG5 FAM115C, CTAGE6, FAM115A, OR2F1, CTAGE4, ARHGEF35, ARHGEF5, NOBOX, TPK1 PDCD1LG2, KIAA1432, ERMP1, MLANA, KIAA2026, RANBP6 UPD: uniparental disomy 5 13q14.11 Deletion FOXO1

17 Supplementary Table 6. Ki67 expression of NKTCL patients according to mutation status. Mutation Status Ki67 50% Ki67 > 50%* P value (mutation vs. wild-type) wild-type DDX3X * TP * One individual with both DDX3X and TP53 mutations was in Ki67 > 50% group and was grouped into two cohorts with mutations.

18 Supplementary Table 7. The upregulated and downregulated genes of NF- B and MAPK pathway in DDX3X-mutated NKTCL revealed by gene expression profiling. GeneSymbol GenbankAccession GenomicCoordinates NF- B pathway PIDD NM_ chr11: IRAK4 NM_ chr12: IL1B NM_ chr2: TRAF2 NM_ chr9: PLCG1 NM_ chr20: ZAP70 NM_ chr2: TNFAIP3 NM_ chr6: PTGS2 NM_ chr1: CXCL2 NM_ chr4: PLAU NM_ chr10: TRAF3 NM_ chr14: VCAM1 NM_ chr1: DDX58 NM_ chr9: NFKBIA NM_ chr14: TRAF1 NM_ chr9: IKBKG NM_ chrx: TRADD NM_ chr16: MAP3K14 NM_ chr17: PIAS4 AK chr19: TAB2 NM_ chr6: RELA NM_ chr11: LTA NM_ chr6: TNF NM_ chr6: TRIM25 NM_ chr17: XIAP NM_ chrx: ERC1 NM_ chr12: CCL4 NM_ chr17: TNFRSF13C NM_ chr22: ICAM1 NM_ chr19: NFKB2 NM_ chr10: LTB NM_ chr6: GADD45B NM_ chr19: CARD11 AK chr7: CCL19 NM_ chr9: LAT NM_ chr16: CD14 NM_ chr5: BLNK NM_ chr10: MYD88 NM_ chr3: BIRC3 NM_ chr11: RELB NM_ chr19: CD40LG NM_ chrx: BTK NM_ chrx: TRAF5 NM_ chr1:

19 CD40 NM_ chr20: RIPK1 NM_ chr6: TNFSF13B NM_ chr13: PLCG2 NM_ chr16: PRKCQ NM_ chr10: TNFRSF1A NM_ chr12: UBE2I XM_ chr16: CHUK NM_ chr10: TLR4 NM_ chr9: TAB3 NM_ chrx: LTBR NM_ chr12: SYK NM_ chr9: LY96 NM_ chr8: BCL2 NM_ chr18: BCL2A1 NM_ chr15: TAB1 NM_ chr22: TRAF6 NM_ chr11: CSNK2A2 NM_ chr16: TICAM1 NM_ chr19: BIRC2 NM_ chr11: TNFSF11 NM_ chr13: PRKCB NM_ chr16: CCL21 NM_ chr9: CCL13 NM_ chr17: CSNK2B NM_ chr6: IL1R1 NM_ chr2: IRAK1 NM_ chrx: LYN NM_ chr8: TNFSF14 NM_ chr19: CXCL12 NM_ chr10: ATM NM_ chr11: CSNK2A1 NM_ chr20: MAP3K7 NM_ chr6: CFLAR NM_ chr2: TIRAP NM_ chr11: IKBKB XM_ chr8: LCK NM_ chr1: NFKB1 NM_ chr4: PARP1 NM_ chr1: MALT1 NM_ chr18: IL8 NM_ chr4: BCL10 NM_ chr1: BCL2L1 NM_ chr20: MAPK pathway FOS NM_ chr14: JUN NM_ chr1: DUSP1 NM_ chr5:

20 DUSP5 NM_ chr10: CACNA1E NM_ chr1: RASGRP1 NM_ chr15: PLA2G2C NM_ chr1: STK3 NM_ chr8: MAP2K3 NM_ chr17: JUND NM_ chr19: CDC25B NM_ chr20: PRKX NM_ chrx: PLA2G2D BC chr1: FGF20 NM_ chr8: GADD45G NM_ chr9: NFKB2 NM_ chr10: JMJD7-PLA2G4B NM_ chr15: NFATC4 NM_ chr14: GADD45B NM_ chr19: MAP2K4 NM_ chr17: MAP3K12 NM_ chr12: DUSP2 NM_ chr2: MAP4K2 NM_ chr11: ELK4 NM_ chr1: MAP3K4 NM_ chr6: CDC42 NM_ chr1: CACNA2D2 NM_ chr3: PDGFB NM_ chr22: MAPK8IP3 NM_ chr16: AKT2 NM_ chr19: MAP3K14 NM_ chr17: HSPA1B NM_ chr6: TRAF2 NM_ chr9: CACNB3 NM_ chr12: MAP3K1 NM_ chr5: IL1B NM_ chr2: MAP3K6 NM_ chr1: DUSP4 NM_ chr8: NR4A1 NM_ chr12: IKBKG NM_ chrx: CD14 NM_ chr5: CACNB2 NM_ chr10: FGFR1 NM_ chr8: MRAS NM_ chr3: MAP4K4 NM_ chr2: FGFR3 NM_ chr4: GNG12 NM_ chr1: ATF2 NM_ chr2: HSPA1A NM_ chr6: MAP3K5 NM_ chr6: MAP4K1 NM_ chr19:

21 PDGFRB NM_ chr5: MAPK10 NM_ chr4: HSPA6 NM_ chr1: NLK NM_ chr17: TAB2 NM_ chr6: TGFB3 NM_ chr14: MKNK1 NM_ chr1: CACNA1A NM_ chr19: HSPB1 NM_ chr7: RELA NM_ chr11: PLA2G6 NM_ chr22: CACNA1C NM_ chr12: BRAF NM_ chr7: FGF2 NM_ chr4: TAOK1 NM_ chr17: MAPK7 NM_ chr17: TNF NM_ chr6: PPP3CC NM_ chr8: RASGRP2 NM_ chr11: RELB NM_ chr19: MAP3K11 NM_ chr11: MAP3K3 NM_ chr17: GADD45A NM_ chr1: PLA2G5 NM_ chr1: PDGFA NM_ chr7: MAP3K8 NM_ chr10: SOS1 NM_ chr2: KRAS NM_ chr12: ATF4 NM_ chr22: AKT1 NM_ chr14: ARRB2 NM_ chr17: FGF7 NM_ chr15: HRAS NM_ chr11: CACNA1B NM_ chr9: STK4 NM_ chr20: RPS6KA2 NM_ chr6: FAS NM_ chr10: TNFRSF1A NM_ chr12: MAPK11 NM_ chr22: NTRK2 NM_ chr9: SRF NM_ chr6: CACNG3 NM_ chr16: MAPKAPK5 NM_ chr12: RASGRP3 NM_ chr2: DUSP6 NM_ chr12: CHUK NM_ chr10: DDIT3 NM_ chr12: RRAS NM_ chr19:

22 CACNA2D4 NM_ chr12: TRAF6 NM_ chr11: TAOK2 NM_ chr16: MEF2C NM_ chr5: RAF1 NM_ chr3: MAPK8 NM_ chr10: CACNA1I NM_ chr22: CACNA1H NM_ chr16: PAK2 NM_ chr3: DUSP16 NM_ chr12: RPS6KA4 NM_ chr11: BDNF-AS NR_ chr11: FLNB NM_ chr3: ECSIT NM_ chr19: MAP3K2 NM_ chr2: MAX NM_ chr14: TGFBR1 NM_ chr9: TAB1 NM_ chr22: NFATC2IP NM_ chr16: MAP2K7 NM_ chr19: FGFR2 NM_ chr10: RAPGEF2 NM_ chr4: FGF11 NM_ chr17: STMN1 NM_ chr1: MYC NM_ chr8: MKNK2 NM_ chr19: RASA1 NM_ chr5: PRKCB NM_ chr16: HSPA2 NM_ chr14: PRKACB NM_ chr1: MAP2K1 NM_ chr15: MAP4K3 NM_ chr2: NRAS NM_ chr1: PRKACA NM_ chr19: TGFBR2 NM_ chr3: MAPKAPK3 NM_ chr3: MAPK9 NM_ chr5: MAPK13 NM_ chr6: MECOM NM_ chr3: RASA2 NM_ chr3: RAP1A NM_ chr1: PPP3R1 NM_ chr2: PDGFRA NM_ chr4: CRK NM_ chr17: IL1R1 NM_ chr2: TAOK3 NM_ chr12: HSPA8 NM_ chr11: PLA2G2A NM_ chr1:

23 IL1R2 NM_ chr2: RRAS2 NM_ chr11: MAP2K5 NM_ chr15: TGFB1 NM_ chr19: RPS6KA5 NM_ chr14: FASLG NM_ chr1: MAPK1 NM_ chr22: DUSP3 NM_ chr17: CRKL NM_ chr22: PLA2G1B NM_ chr12: PTPN7 NM_ chr1: PPP3CA NM_ chr4: RASGRF1 NM_ chr15: RAC3 NM_ chr17: GRB2 NM_ chr17: PLA2G12A NM_ chr4: LAMTOR3 NM_ chr4: RPS6KA1 NM_ chr1: RAP1B NM_ chr12: RAC1 NM_ chr7: DUSP14 NM_ chr17: MAP2K6 NM_ chr17: SOS2 NM_ chr14: PLA2G4A NM_ chr1: ZAK NM_ chr2: EGFR NM_ chr7: PPM1B NM_ chr2: FGF1 NM_ chr5: MAPK3 NM_ chr16: DAXX NM_ chr6: RPS6KA3 NM_ chrx: MAP3K7 NM_ chr6: FGF14-AS2 NR_ chr13: RAC2 NM_ chr22: FLNA NM_ chrx: FGF13 NM_ chrx: MAPKAPK2 NM_ chr1: GNA12 NM_ chr7: CASP3 NM_ chr4: DUSP10 NM_ chr1: NF1 NM_ chr17: ARRB1 NM_ chr11: PPP3CB AK chr10: HSPA1L NM_ chr6: MAPK8IP2 AK chr22: IKBKB XM_ chr8: NFKB1 NM_ chr4: PAK1 TCONS_ chr11:

24 MAP2K2 NM_ chr19: PRKCA NM_ chr17: AKT3 NM_ chr1: DUSP8 NM_ chr11: MOS NM_ chr8: RASGRP4 NM_ chr19:

25 Supplementary Table 8. Clinical characteristics of sequencing and validation cohorts of NKTCL patients with follow-up data (n = 95). Variable wild-type (n = 62) DDX3X (n = 20) mutations* TP53 (n = 14) Age (years) Gender Ann Arbor stage ECOG Extranodal involvement LDH level IPI score B symptom Therapy > P = P = Female Male P = P = I II III IV P = P = P = P = < P = P = Normal Elevated Low and low-intermediate risk (0 2) High-intermediate and high risk (3 5) P = P = P = P = Absent Present P = P = Chemotherapy alone Chemotherapy plus Radiotherapy P = P = * One individual with both DDX3X and TP53 mutations was grouped into two cohorts with mutations. LDH, lactate dehydrogenase; IPI, International Prognostic Index. All variables were analyzed by Fisher s exact test. The P value was obtained by comparing wild-type and DDX3X mutations groups or DDX3X and TP53 mutations groups.

26 Supplementary Table 9. The multivariate analysis and univariate analysis for clinical and biological characteristics of overall survival and progression-free survival in NKTCL patients (n = 95*). Variable Multivariate analysis OS PFS P OR (95% CI) P OR (95% CI) DDX3X (mutation vs. wild-type) ( ) ( ) TP (mutation vs. wild-type) ( ) ( ) IPI score (3 5 vs. 0 2) ( ) ( ) B symptom (present vs. absent) NS Radiotherapy (yes vs. no) ( ) Extranasal involvement (yes vs. no) NS - Univariate analysis Variable P (OS) P (PFS) NS ( ) Age (> 60y vs. 60y) Gender (male vs. female) DDX3X (mutation vs. wild-type) TP53 (mutation vs. wild-type) IPI score (3 5 vs. 0 2) B symptom (present vs. absent) Radiotherapy (yes vs. no) Extranasal involvement (yes vs. no) * OS, overall survival; PFS, progression free survival; IPI, International Prognostic Index; wild-type, individuals without DDX3X or TP53 mutations; mutation, individuals with DDX3X or TP53 mutations.

27 Supplementary Table 10. Sequencing analysis summary of samples from 25 NKTCL patients subject to whole-exome sequencing. Patient ID L01 L02 L03 L04 L05 L06 L07 L08 L09 L10 L11 L12 L13 L14 L15 L16 L17 L18 L19 L20 L21 Sample Depth * Coverage * 1X 2X 4X 8X 10X 15X 20X Tumor % 93.3% 90.2% 85.9% 84.1% 79.8% 76.1% Blood % 92.5% 89.2% 84.5% 82.5% 78.0% 74.2% Tumor % 95.2% 92.8% 88.4% 86.3% 81.0% 76.0% Blood % 95.2% 92.9% 88.9% 87.0% 82.3% 77.9% Tumor % 94.9% 92.8% 89.0% 87.0% 82.1% 77.2% Blood % 95.4% 93.2% 89.4% 87.5% 82.7% 78.1% Tumor % 94.8% 92.2% 88.4% 86.7% 83.0% 79.4% Blood % 94.5% 91.8% 87.9% 86.2% 82.3% 78.6% Tumor % 94.5% 91.9% 88.0% 86.3% 82.3% 78.6% Blood % 94.8% 92.2% 88.4% 86.9% 83.2% 79.7% Tumor % 94.8% 92.4% 88.8% 87.3% 83.6% 80.1% Blood % 94.3% 91.8% 87.9% 86.2% 82.3% 78.6% Tumor % 95.1% 93.0% 89.2% 87.4% 82.6% 78.0% Blood % 94.9% 92.8% 89.1% 87.3% 82.9% 78.6% Tumor % 94.5% 91.9% 87.4% 85.2% 79.9% 74.8% Blood % 94.5% 91.6% 86.5% 84.0% 77.6% 71.5% Tumor % 95.3% 93.2% 89.6% 87.8% 83.3% 78.8% Blood % 95.1% 93.1% 89.5% 87.9% 83.7% 79.5% Tumor % 93.2% 89.9% 85.2% 83.0% 78.1% 73.8% Blood % 92.3% 88.7% 83.5% 81.2% 76.1% 71.8% Tumor % 93.1% 89.8% 84.8% 82.6% 77.6% 73.2% Blood % 93.0% 89.6% 84.5% 82.3% 77.1% 72.7% Tumor % 93.9% 91.0% 86.7% 84.8% 80.3% 76.2% Blood % 94.5% 91.8% 87.5% 85.6% 80.6% 75.6% Tumor % 93.4% 90.0% 84.7% 82.3% 76.5% 71.3% Blood % 95.6% 93.3% 89.6% 88.0% 84.1% 80.0% Tumor % 97.6% 96.7% 95.3% 94.6% 93.1% 91.7% Blood % 97.6% 96.6% 94.9% 94.2% 92.5% 91.0% Tumor % 98.6% 98.3% 97.9% 97.7% 97.2% 96.8% Blood % 98.6% 98.3% 97.8% 97.6% 97.2% 96.8% Tumor % 97.8% 96.8% 95.3% 94.6% 93.0% 91.6% Blood % 97.4% 96.3% 94.4% 93.5% 91.7% 90.0% Tumor % 98.8% 98.5% 98.0% 97.8% 97.3% 96.8% Blood % 98.7% 98.3% 97.8% 97.6% 97.0% 96.4% Tumor % 98.3% 97.8% 97.0% 96.6% 95.5% 94.1% Blood % 98.6% 98.2% 97.7% 97.5% 97.0% 96.4% Tumor % 98.6% 98.2% 97.5% 97.3% 96.5% 95.7% Blood % 98.7% 98.3% 97.7% 97.5% 96.8% 96.0% Tumor % 98.7% 98.4% 97.9% 97.6% 97.0% 96.4% Blood % 98.6% 98.3% 97.8% 97.5% 96.9% 96.3% Tumor % 98.5% 98.1% 97.6% 97.3% 96.6% 95.9% Blood % 98.6% 98.2% 97.7% 97.4% 96.8% 96.2%

28 L22 L23 L24 Tumor % 98.6% 98.2% 97.6% 97.3% 96.5% 95.6% Blood % 98.8% 98.4% 97.9% 97.6% 97.0% 96.4% Tumor % 98.4% 98.0% 97.2% 96.9% 95.8% 94.6% Blood % 98.4% 98.0% 97.3% 96.9% 96.0% 94.8% Tumor % 98.9% 98.1% 97.0% 96.5% 95.1% 93.3% Blood % 98.6% 98.3% 97.7% 97.5% 96.9% 96.3% Tumor % 98.4% 98.0% 97.3% 97.0% 96.0% 94.9% L25 Blood % 98.5% 98.0% 97.4% 97.0% 96.1% 95.0% * Based on 33Mb RefSeq coding sequence region.

29 Supplementary Table 11. The RNA oligonucleotide sequences of RNA unwinding assay. Name Sequence (5 3 ) r18mer_fam r36mer CCCAAGAACCCAAGGAAC ACCAGCUUUGUUCCUUGGGUUCUUGGGAGCAGCAGG

30 Supplementary Table 12. The primer sequences of DDX3X shrna. Name Sequence (5 3 ) DDX3X-shRNA1 DDX3X-shRNA2 Non-targeting shrna GATCCGGATCTCGTAGTGATTCAAGATTCAAGAGATCTTGAATCACTACGAGAT CCTTTTTTACGCGTG GATCCGGTAGAATAGTCGAACAAGATTTCAAGAGAATCTTGTTCGACTATTCTAC CTTTTTTACGCGTG GATCCAGCGTGTAGCTAGCAGAGGTTCAAGAGACCTCTGCTAGCTACACGCTT TTTTTACGCGTG