Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine D i ff e re n t i a l E x p re s s i o n – B re a s t v s L u n g  Work Log 2/1 

Datasets Summary Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 2 Dataset ID Type # Sample Platform Description GSE39162 (A) Breast 15 (paired T, TN, N) GA, GAII de-novo N type not used GSE33858 (B) Lung 32 (paired T, NT) GAIIx GSE29173 (C) Breast 245 (unpaired, #Normal = 16) GAIIx barcoded 

Expression Difference in Breast and Lung Cancer •  A+C = breast cancer, B = lung cancer •  Welch’s t test for finding expression difference between breast and lung cancer •  unequal variance in two type, Behrens–Fisher problem. •  type of samples(#) used •  A: T(5), B: T(16), C: not Normal(229) •  Negative test •  do t test on normal type of samples of both cancer types. •  type of samples(#) used: •  A: TN (N excluded)(5), B: NT(16), C: Normal(16) •  if the expression difference is significant, then we should question whether they are all “normal” samples. Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 3 

Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 4 0 10000 20000 30000 40000 50000 0 10000 20000 30000 40000 50000 breast lung chr7_8791 chr11_13342 chr22_20736 chr13_14817 chr17_17828 chr20_19494 chr20_19450 chr18_18769 chr11_13709 chr2_2356 chr4_5692 chr11_12760 chr3_3910 chr1_692 chr10_12452 chr6_7548 chr1_944 chr6_8151 chr11_13239 chr7_8991 chr7_8849 chr22_20809 chr2_3487 chr7_8673 chr14_15459 chr20_19463 chr6_8250 chr19_19992 chr1_689 chr17_17785 Expression of miRNA candidates in breast and lung cancer datasets (p = 0.05) Reads Per Million 

Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 5 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 breast lung chr7_8791 chr11_13342 chr22_20736 chr13_14817 chr17_17828 chr20_19494 chr20_19450 chr18_18769 chr11_13709 chr2_2356 chr4_5692 chr11_12760 chr3_3910 chr1_692 chr10_12452 chr6_7548 chr1_944 chr6_8151 chr11_13239 chr7_8991 chr7_8849 chr22_20809 chr2_3487 chr7_8673 chr14_15459 chr20_19463 chr6_8250 chr19_19992 chr1_689 chr17_17785 Expression of miRNA candidates in breast and lung cancer datasets (p = 0.05) Reads Per Million many candidates having median at 0 (many zero expression samples)  

Result Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 6 breast- mean lung- mean p- value Negative test chr7_8791 0.250 0.077 0.047 chr11_13342 0.569 0.177 0.047 chr22_20736 0.250 0.162 0.379 chr13_14817 0.014 0.003 0.275 chr17_17828 0.329 0.000 0.000 chr20_19494 0.084 0.032 0.328 chr20_19450 0.104 0.018 0.129 chr18_18769 0.348 0.000 0.005 chr11_13709 0.126 0.000 0.275 chr2_2356 0.390 0.382 0.955 chr4_5692 0.428 0.080 0.130 chr11_12760 0.060 0.037 0.659 chr3_3910 0.571 0.000 0.016 not passed chr1_692 0.238 0.005 0.000 chr10_12452 0.049 0.017 0.246 breast- mean lung- mean p-value Negative test chr6_7548 0.343 0.233 0.436 chr1_944 0.053 0.002 0.287 chr6_8151 0.340 0.233 0.450 chr11_13239 0.070 0.157 0.182 chr7_8991 0.149 0.037 0.120 chr7_8849 0.451 0.041 0.004 chr22_20809 305.666 3.200 0.000 not passed chr2_3487 10.522 0.018 0.000 not passed chr7_8673 0.415 0.004 0.124 chr14_15459 0.954 0.236 0.000 chr20_19463 0.117 0.023 0.158 chr6_8250 4715.337 0.060 0.000 not passed chr19_19992 11225.071 62.053 0.000 not passed chr1_689 0.033 0.000 0.122 chr17_17785 301.149 0.355 0.000 not passed 

Why some candidates don’t pass negative check? Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 7 0 10000 20000 30000 40000 50000 0 10000 20000 30000 40000 50000 0 10000 20000 30000 40000 50000 GSE39162(A) breast GSE33858(B) lung GSE29173(C) breast chr7_8791 chr11_13342 chr22_20736 chr13_14817 chr17_17828 chr20_19494 chr20_19450 chr18_18769 chr11_13709 chr2_2356 chr4_5692 chr11_12760 chr3_3910 chr1_692 chr10_12452 chr6_7548 chr1_944 chr6_8151 chr11_13239 chr7_8991 chr7_8849 chr22_20809 chr2_3487 chr7_8673 chr14_15459 chr20_19463 chr6_8250 chr19_19992 chr1_689 chr17_17785 Expression of normal type in 3 datasets Reads Per Million 

Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 8 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 2.5 GSE39162(A) breast GSE33858(B) lung GSE29173(C) breast chr7_8791 chr11_13342 chr22_20736 chr13_14817 chr17_17828 chr20_19494 chr20_19450 chr18_18769 chr11_13709 chr2_2356 chr4_5692 chr11_12760 chr3_3910 chr1_692 chr10_12452 chr6_7548 chr1_944 chr6_8151 chr11_13239 chr7_8991 chr7_8849 chr22_20809 chr2_3487 chr7_8673 chr14_15459 chr20_19463 chr6_8250 chr19_19992 chr1_689 chr17_17785 Expression of normal type in 3 datasets Reads Per Million •  Candidates with median at 0 (most pink) does not mean they are all 0 values, but having large variance. •  samples from C have overall larger variance 

Candidates having significant diff. exp. Bioinformatics and Biostatistics Core, NTU Center of Genomic Medicine 9 ID location gene gene function chr7_8791 intron ZYX zyxin chr11_13342 exon BTBD10 糖代謝 chr17_17828 intron CASC3 cancer susceptibility candidate chr18_18769 intergenic - chr3_3910 intron NR2C2 chr1_692 intron CREB3L4 cAMP related chr7_8849 3'UTR RBM33 RNA binding motif chr22_20809 intron AP1B1P1 pseudogene chr2_3487 intergenic - chr14_15459 intron IF127 interferon chr6_8250 intron TULP4 tubby like protein chr19_19992 intron SPTBN4 beta-spectrin chr17_17785 inron ACCN1 DEG/ENaC, neurotransmission, Multiple Sclerosis red: pass negative test