Blue:RNA-Seq Analyses:PBMCs:20140523 PBMCs: Difference between revisions

From ZhangLabWiki
Jump to navigation Jump to search
>B1lake
>B1lake
Line 516: Line 516:
|}
|}


== Bicluster Analysis: Separation of two Groups ==
== Bicluster Analysis: Splitting Samples into two Groups ==


*To try and separate out single PBMC groupings prior to Singular analysis (requires comparison of at least two data sets) without using human Neurons
*To try and separate out single PBMC groupings prior to Singular analysis (requires comparison of at least two data sets) without using human Neurons

Revision as of 19:14, 28 October 2014

Libraries

Group Sample Name Nextera_Ind Seq Lane Seq Run Seq File 1 Seq File 2 Seq File 3
1 1_PBMC_C03 1 3,4,5 HL170 s_3_1_Indx01.txt.gz s_4_1_Indx01.txt.gz s_5_1_Indx01.txt.gz
1 1_PBMC_C09 2 3,4,5 HL170 s_3_1_Indx02.txt.gz s_4_1_Indx02.txt.gz s_5_1_Indx02.txt.gz
1 1_PBMC_C15 3 3,4,5 HL170 s_3_1_Indx03.txt.gz s_4_1_Indx03.txt.gz s_5_1_Indx03.txt.gz
1 1_PBMC_C25 4 3,4,5 HL170 s_3_1_Indx04.txt.gz s_4_1_Indx04.txt.gz s_5_1_Indx04.txt.gz
1 1_PBMC_C43 5 3,4,5 HL170 s_3_1_Indx05.txt.gz s_4_1_Indx05.txt.gz s_5_1_Indx05.txt.gz
1 1_PBMC_C14 6 3,4,5 HL170 s_3_1_Indx06.txt.gz s_4_1_Indx06.txt.gz s_5_1_Indx06.txt.gz
1 1_PBMC_C20 7 3,4,5 HL170 s_3_1_Indx07.txt.gz s_4_1_Indx07.txt.gz s_5_1_Indx07.txt.gz
1 1_PBMC_C26 8 3,4,5 HL170 s_3_1_Indx08.txt.gz s_4_1_Indx08.txt.gz s_5_1_Indx08.txt.gz
1 1_PBMC_C01 9 3,4,5 HL170 s_3_1_Indx09.txt.gz s_4_1_Indx09.txt.gz s_5_1_Indx09.txt.gz
1 1_PBMC_C39 10 3,4,5 HL170 s_3_1_Indx10.txt.gz s_4_1_Indx10.txt.gz s_5_1_Indx10.txt.gz
1 1_PBMC_C61 11 3,4,5 HL170 s_3_1_Indx11.txt.gz s_4_1_Indx11.txt.gz s_5_1_Indx11.txt.gz
1 1_PBMC_C87 12 3,4,5 HL170 s_3_1_Indx12.txt.gz s_4_1_Indx12.txt.gz s_5_1_Indx12.txt.gz
1 1_PBMC_C56 13 3,4,5 HL170 s_3_1_Indx13.txt.gz s_4_1_Indx13.txt.gz s_5_1_Indx13.txt.gz
1 1_PBMC_C68 14 3,4,5 HL170 s_3_1_Indx14.txt.gz s_4_1_Indx14.txt.gz s_5_1_Indx14.txt.gz
1 1_PBMC_C74 15 3,4,5 HL170 s_3_1_Indx15.txt.gz s_4_1_Indx15.txt.gz s_5_1_Indx15.txt.gz
1 1_PBMC_C80 16 3,4,5 HL170 s_3_1_Indx16.txt.gz s_4_1_Indx16.txt.gz s_5_1_Indx16.txt.gz
1 1_PBMC_C92 17 3,4,5 HL170 s_3_1_Indx17.txt.gz s_4_1_Indx17.txt.gz s_5_1_Indx17.txt.gz
1 1_PBMC_C51 18 3,4,5 HL170 s_3_1_Indx18.txt.gz s_4_1_Indx18.txt.gz s_5_1_Indx18.txt.gz
1 1_PBMC_C57 19 3,4,5 HL170 s_3_1_Indx19.txt.gz s_4_1_Indx19.txt.gz s_5_1_Indx19.txt.gz
1 1_PBMC_C63 20 3,4,5 HL170 s_3_1_Indx20.txt.gz s_4_1_Indx20.txt.gz s_5_1_Indx20.txt.gz
2 1_PBMC_C69 1 6,7,8 HL170 s_6_1_Indx01.txt.gz s_7_1_Indx01.txt.gz s_8_1_Indx01.txt.gz
2 1_PBMC_C06 2 6,7,8 HL170 s_6_1_Indx02.txt.gz s_7_1_Indx02.txt.gz s_8_1_Indx02.txt.gz
2 1_PBMC_C18 3 6,7,8 HL170 s_6_1_Indx03.txt.gz s_7_1_Indx03.txt.gz s_8_1_Indx03.txt.gz
2 1_PBMC_C24 4 6,7,8 HL170 s_6_1_Indx04.txt.gz s_7_1_Indx04.txt.gz s_8_1_Indx04.txt.gz
2 1_PBMC_C28 5 6,7,8 HL170 s_6_1_Indx05.txt.gz s_7_1_Indx05.txt.gz s_8_1_Indx05.txt.gz
2 1_PBMC_C34 6 6,7,8 HL170 s_6_1_Indx06.txt.gz s_7_1_Indx06.txt.gz s_8_1_Indx06.txt.gz
2 1_PBMC_C46 7 6,7,8 HL170 s_6_1_Indx07.txt.gz s_7_1_Indx07.txt.gz s_8_1_Indx07.txt.gz
2 1_PBMC_C17 8 6,7,8 HL170 s_6_1_Indx08.txt.gz s_7_1_Indx08.txt.gz s_8_1_Indx08.txt.gz
2 1_PBMC_C35 9 6,7,8 HL170 s_6_1_Indx09.txt.gz s_7_1_Indx09.txt.gz s_8_1_Indx09.txt.gz
2 1_PBMC_C41 10 6,7,8 HL170 s_6_1_Indx10.txt.gz s_7_1_Indx10.txt.gz s_8_1_Indx10.txt.gz
2 1_PBMC_C04 11 6,7,8 HL170 s_6_1_Indx11.txt.gz s_7_1_Indx11.txt.gz s_8_1_Indx11.txt.gz
2 1_PBMC_C48 12 6,7,8 HL170 s_6_1_Indx12.txt.gz s_7_1_Indx12.txt.gz s_8_1_Indx12.txt.gz
2 1_PBMC_C64 13 6,7,8 HL170 s_6_1_Indx13.txt.gz s_7_1_Indx13.txt.gz s_8_1_Indx13.txt.gz
2 1_PBMC_C78 14 6,7,8 HL170 s_6_1_Indx14.txt.gz s_7_1_Indx14.txt.gz s_8_1_Indx14.txt.gz
2 1_PBMC_C90 15 6,7,8 HL170 s_6_1_Indx15.txt.gz s_7_1_Indx15.txt.gz s_8_1_Indx15.txt.gz
2 1_PBMC_C53 16 6,7,8 HL170 s_6_1_Indx16.txt.gz s_7_1_Indx16.txt.gz s_8_1_Indx16.txt.gz
2 1_PBMC_C65 17 6,7,8 HL170 s_6_1_Indx17.txt.gz s_7_1_Indx17.txt.gz s_8_1_Indx17.txt.gz
2 1_PBMC_C54 18 6,7,8 HL170 s_6_1_Indx18.txt.gz s_7_1_Indx18.txt.gz s_8_1_Indx18.txt.gz
2 1_PBMC_C88 19 6,7,8 HL170 s_6_1_Indx19.txt.gz s_7_1_Indx19.txt.gz s_8_1_Indx19.txt.gz
2 1_PBMC_C94 20 6,7,8 HL170 s_6_1_Indx20.txt.gz s_7_1_Indx20.txt.gz s_8_1_Indx20.txt.gz
TubeCont CD4+ 21 1 HL170 s_1_1_Indx21.txt.gz
TubeCont CD19+ 22 1 HL170 s_1_1_Indx22.txt.gz
TubeCont PBMC 23 1 HL170 s_1_1_Indx23.txt.gz

Mapping Statistics

  • TSCC Mapping:
batch_STAR_cufflink2_HTseq.pl hg19
  • Mapping analysis in Genome Miner:
/home/kunzhang/RNAseq/SCAP/scripts/get_STAR_mapping_stats.pl PBMC_HL170 > PBMC_HL170_mapping_stats.txt
/home/kunzhang/RNAseq/SCAP/scripts/get_STAR_TPM_matrix.pl PBMC_HL170 > PBMC_HL170_TPM.txt
  • Summary:

File:Hg19 ERCC Ratio.jpg

File:Mapping Stats.jpg

File:Mapping Types.jpg

Singular Analysis: Comparison with Neurons

  • Neuron libraries: C1 huNu experiment 20140227
  • Removed unannotated genes (i.e. all genes without HGNC Names)
  • Singular Software 3.0 AutoAnalysis: LOD = 1; Top 100 differentially expressed genes


Principal Component Analysis:

File:PBMC Neuron PCA Plot.png


Hierarchical Clustering:


File:PBMC Neuron HC Plot.png

  • Can identify two potential subpoulations of PBMCs (labeled Group1 and Group2)
  • Can identify three subsets of PBMC-related genes (Set1, Set2, Set3):
GeneID GroupID ' GeneID GroupID ' GeneID GroupID
SMCHD1 Set1 FAM208B Set2 MAP3K1 Set3
PHC3 Set1 HIVEP2 Set2 HLA-DRA Set3
PDE7A Set1 HSP90AB1 Set2 CD74 Set3
WAC Set1 ANK3 Set2 LYN Set3
RPL3 Set1 TESPA1 Set2 ARHGAP24 Set3
STK17B Set1 ABLIM1 Set2 NEAT1 Set3
RPL30 Set1 OXNAD1 Set2 AOAH Set3
LAPTM5 Set1 CAMK4 Set2 FGR Set3
RASSF3 Set1 PRKCA Set2 MYO1F Set3
DOCK2 Set1 PDK1 Set2 RBMS1 Set3
ZBTB20 Set1 IL6ST Set2 GNAQ Set3
FYB Set1 BCL11B Set2 HIF1A-AS2 Set3
PTPRC Set1 LRRC8C Set2 ERBB2IP Set3
B2M Set1 LRRC75A Set2 PPP1R12A Set3
LRRC75A-AS1 Set2 VAPA Set3
SKP1 Set2 NCOA1 Set3
TC2N Set2 LRRFIP1 Set3
CCR7 Set2 OSBPL8 Set3
TRAC Set2 CAST Set3
PCED1B Set2 SMG7 Set3
MAML2 Set2 NCF2 Set3
CD96 Set2 C10orf11 Set3
ITK Set2 RBM47 Set3
IL7R Set2 DOCK5 Set3
LDHB Set2 LYZ Set3
CD3G Set2 VCAN Set3
CD3D Set2 CPVL Set3
LEF1 Set2 MAML3 Set3
PLXDC2 Set3
VMP1 Set3
RAP1GAP2 Set3
EVI5 Set3
ZEB2 Set3
MARCH1 Set3
JAZF1 Set3


Violin Plot


File:PBMC Neuron Violin Plot.png

  • PBMC Upregulated Genes (Log2Ex Values >2fold)
ID PBMC Neuron
B2M 12.06088994 0
PTPRC 11.43596383 1.246360079
MAML2 10.74831794 3.139034281
FYB 10.47348499 0
ZBTB20 10.07262771 4.365163912
IL7R 9.958356597 0
LEF1 9.924274838 0
STK17B 9.712558873 0
DOCK2 9.30727295 2.349849452
PDE7A 9.165049555 4.056165937
MAP3K1 9.135716522 2.156183693
ERBB2IP 8.885414539 4.439420173
PCED1B 8.831443091 3.733026029
VMP1 8.759459637 4.360635962
AOAH 8.70815656 2.212552957
TC2N 8.668049838 0.38107932
HLA-DRA 8.631302971 0
LDHB 8.578374713 2.815661712
CD96 8.56703784 0
ITK 8.533862961 0
RPL30 8.485380678 1.327596607
ARHGAP24 8.425139607 1.330694255
CD3G 8.383179076 0
RASSF3 8.310063349 0
CPVL 8.206916149 3.184292824
CAST 8.190390515 4.027333343
IL6ST 8.001360942 3.384831455
SKP1 7.841030974 3.52743585
VCAN 7.837267496 0.283828219
LRRC8C 7.812971223 3.532507267
RPL3 7.713171439 1.899806557
LYZ 7.670636524 0
PDK1 7.60225638 2.734495713
CD74 7.341776762 0
LYN 7.329245674 0
TRAC 7.074289051 0
LAPTM5 7.068575619 0
C10orf11 7.03026631 2.820673439
RBM47 6.9559114 0.450124234
HIF1A-AS2 6.852078227 2.774368163
NCF2 6.742711713 0
LRRC75A-AS1 6.624588689 2.261847491
CCR7 6.457177404 0
FGR 6.302947632 0
DOCK5 6.259374053 0.49461772
CD3D 6.096667723 0
LRRC75A 6.018687421 2.076496088
MYO1F 5.785705604 0.622264566

Singular Analysis: Comparison within Groups

  • Comparison of putative PBMC sub-groups identified above (see PBMC Neuron Hierarchical Cluster)
  • Singular 3.0: Autoanalysis: LOD1; top 100 differentially expressed genes


PCA Analysis: Group 1/2

  • 3D PCA Plot

File:PBMC Group 3d PCA plot.png


Hierarchical Clustering Analysis: Group 1/2

File:PBMC Group HC Plot.png


  • Note that there remains heterogeneity in these groups, but further splitting is impossible due to the low sample sizes
  • Can identify 4 potential gene clusters (Set1, Set2, Set3, Set4):
GeneID GroupID ' GeneID GroupID ' GeneID GroupID ' GeneID GroupID
MYBL1 Set 1 MARCH1 Set 2 RICTOR Set 3 LCLAT1 Set 4
ID2 Set 1 CD74 Set 2 STK17B Set 3 INPP4B Set 4
NASP Set 1 JAZF1 Set 2 STRBP Set 3 PPIL4 Set 4
VMP1 Set 1 SETBP1 Set 2 FAM117B Set 3 FHIT Set 4
ZFAND6 Set 1 CCDC88A Set 2 RALGPS2 Set 3 MAML2 Set 4
QKI Set 1 HDAC9 Set 2 BACH2 Set 3 OXNAD1 Set 4
AOAH Set 1 CAST Set 2 CGGBP1 Set 3 PLCL1 Set 4
FGR Set 1 SMG7 Set 2 RNF111 Set 3 BCLAF1 Set 4
ZEB2 Set 1 APLP2 Set 2 RNF2 Set 3 LRRC75A Set 4
DIAPH2 Set 2 AFF3 Set 3 IL7R Set 4
VCAN-AS1 Set 2 CYLD Set 3 TC2N Set 4
NCF2 Set 2 C1orf112 Set 3 ANK3 Set 4
IRAK3 Set 2 UBA3 Set 3 HIVEP2 Set 4
LYN Set 2 FCHSD2 Set 3 PHC3 Set 4
EVI5 Set 2 R3HDM2 Set 3 TXK Set 4
ARHGAP24 Set 2 WDFY4 Set 3 PRKCQ Set 4
RBM47 Set 2 ADAM28 Set 3 RNF169 Set 4
SLC8A1 Set 2 SFT2D1 Set 3 XRRA1 Set 4
CPVL Set 2 CASP1 Set 3 RPL3 Set 4
C10orf11 Set 2 SESTD1 Set 3 RPL35A Set 4
LYZ Set 2 HLA-DRA Set 3 RPL30 Set 4
DOCK5 Set 2 OSBPL10 Set 3 RPL5 Set 4
MAML3 Set 2 RASGRP3 Set 3 LRRC8C Set 4
VCAN Set 2 BANK1 Set 3 GNB2L1 Set 4
PLXDC2 Set 2 BLK Set 3 LDHB Set 4
MS4A1 Set 3 CD3D Set 4
BCL11A Set 3 CD3G Set 4
EBF1 Set 3 TESPA1 Set 4
CCR7 Set 4
LEF1 Set 4
CAMK4 Set 4
PRKCA Set 4


Violin Plot: Group 1/2

File:PBMC Group Violin Plot.png


  • Top Genes Expressed in Group 1 versus Group 2 (log2Ex > 2):
Top Genes Group 1 ' ' ' Top Genes Group 2 ' '
ID Group1 Group2 ID Group1 Group2
LYN 9.129954615 0.049282249 LEF1 0.555570778 10.40907802
OSBPL10 9.323622829 0.273157782 CAMK4 0 9.597364962
LYZ 9.470710979 0.521905042 PRKCA 1.029187823 9.586852639
MARCH1 9.335855359 0.622883509 TC2N 0 9.152839755
SLC8A1 9.592184637 0.844184763 LDHB 0 9.06354959
MS4A1 11.29801193 0 CD3G 0 8.868453579
PLXDC2 10.31543618 0.963511116 CCR7 0 6.941098092
HLA-DRA 10.43849208 0 CD3D 0 6.58209455
ARHGAP24 10.23188851 0 LRRC75A 0 6.499326104
CPVL 10.01351303 0 TESPA1 1.380452407 8.235450163
ADAM28 9.785255259 0 LCLAT1 1.48864354 7.278017747
VCAN 9.643546202 0 RNF169 1.673443356 7.641024287
EBF1 9.106720614 0 ANK3 2.05883598 9.336394755
BCL11A 8.992256824 0 LRRC8C 1.883669687 8.291618061
SESTD1 8.960251413 0 XRRA1 1.796026626 7.766035558
C10orf11 8.829801606 0 IL7R 2.984612134 10.44050025
MAML3 8.800819433 0 INPP4B 2.818518269 9.434072621
RBM47 8.756778547 0
VCAN-AS1 8.675922641 0
WDFY4 8.09846509 0.99494171
DOCK5 8.063963957 0
BLK 7.816216488 0
SFT2D1 7.233426745 0
RASGRP3 7.055756928 0
BANK1 10.23475098 1.768650147
FGR 8.074886707 1.422475224
CD74 9.118670542 2.246347277
IRAK3 8.334750938 2.694802207
CASP1 8.932061541 3.062657096
EVI5 9.334982725 3.229757099
JAZF1 9.467325084 3.619718186

Bicluster Analysis: Splitting Samples into two Groups

  • To try and separate out single PBMC groupings prior to Singular analysis (requires comparison of at least two data sets) without using human Neurons
  • Wanted to do biclustering to generate a first level hierarchical spit of the samples as per Rui's analysis

Overdispersed genes

/home/kunzhang/RNAseq/SCAP/scripts/exprMatrix2overDispersedGenes.pl PBMC_HL170_TPM.txt > PBMC_HL170_TPM.txt_OverDisp.txt
/home/kunzhang/RNAseq/SCAP/scripts/extract_gene_set_ENSG.pl PBMC_HL170_TPM.txt_OverDisp.txt PBMC_HL170_TPM.txt > PBMC_HL170_OverDisp_TPM_Matrix.txt
/home/kunzhang/RNAseq/SCAP/scripts/filterExprMatrix.pl PBMC_noTC_HL170_TPM.txt > PBMC_noTC_Filtered_HL170_TPM.txt	
  • 28 samples
  • 3096 genes
/home/kunzhang/RNAseq/SCAP/scripts/exprMatrix2overDispersedGenes.pl PBMC_noTC_Filtered_HL170_TPM.txt > PBMC_noTC_Filtered_HL170_TPM_OverDisp.txt	
  • 92 Over-dispersed genes
/home/kunzhang/RNAseq/SCAP/scripts/extract_gene_set_ENSG.pl PBMC_noTC_Filtered_HL170_TPM_OverDisp.txt PBMC_noTC_Filtered_HL170_TPM.txt > PBMC_HL170_Filtered_noTC_OverDisp_TPM_Matrix.txt	


Hierarchical Clustering

library(gplots)
x=read.table("PBMC_HL170_Filtered_noTC_OverDisp_TPM_Matrix.txt",header=TRUE,row.names=1);
x.cor = cor(log10(x+1),use="pairwise.complete.obs",method="pearson")
heatmap.2(as.matrix(x.cor), col=bluered(128), scale="none", cexCol=0.6, cexRow=0.6 ,key=T, symkey=F,density.info="histogram",trace="none",dendrogram="both",Rowv=TRUE,Colv=TRUE)

File:PBMC BiCluster noTC.png


Cutting Cluster Dendrogram

hc.rows<- hclust(dist(x.cor))
plot(hc.rows, cex = 0.6, cex.main = 2)
ct<- cutree(hc.rows, h=3.0)                                                               
rect.hclust(hc.rows, h=3.0) # draw red rectangles to mark the subgroups                                                                 
write.table(ct, file="PBMC_cutree_2_clusters.txt", sep = "\t", row.names=TRUE, col.names=TRUE)  

File:Cutree Dendrogram PBMC.png


Cluster Comparison using Singular 3.0

  • AutoAnalysis using LOD1 and top 100 differentially expressed genes between BiCluster1 and BiCluster2 identified above