-
Notifications
You must be signed in to change notification settings - Fork 13
/
CaRpools-extended-PDF.Rmd
1472 lines (1013 loc) · 69.9 KB
/
CaRpools-extended-PDF.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
---
title: "CaRpools Report - CRISPR Screen Analysis"
output:
pdf_document:
fig_height: 6
fig_width: 11
highlight: null
keep_tex: yes
latex_engine: xelatex
toc: yes
toc_depth: 3
---
```{r loadlibs, echo=FALSE, error=FALSE, warning=FALSE, message=FALSE}
if("caRpools" %in% rownames(installed.packages()) == FALSE) {install.packages("caRpools")}
library(caRpools,warn.conflicts = FALSE, quietly = TRUE,verbose =FALSE)
load.packages()
```
\newpage
\begin{center}
\includegraphics{CaRpools.png}
\end{center}
```{r settings-setup, echo=FALSE, error=FALSE, warning=FALSE, message=FALSE}
#OPTIONAL PROXY
#options(RCurlOptions=list(proxy="YOURproxyHERE", http.version=1))
### Load PARAMETER FILE
# This name can be changed to your needs
if(!exists("miaccs.file"))
{miaccs.file = "MIACCS.xls"}
miaccs = load.file(miaccs.file, type="xlsx")
#### global options to produce the report
knitr::opts_chunk$set(sanitize=TRUE)
knitr::opts_chunk$set(dev=miaccs["carpools.device",3]) # ImageMagick needs to be installed!
knitr::opts_chunk$set(dpi=as.numeric(miaccs["carpools.dpi",3]))
knitr::opts_chunk$set(results="asis")
options(scipen = 0)
#### Screen Information
screen.id = as.character(miaccs["screening.id",3])
screen.date = as.character(miaccs["screening.date",3])
screen.hypothesis = as.character(miaccs["hypothesis",3])
screen.description = as.character(miaccs["assay.description",3])
screen.organism = as.character(miaccs["assay.organism",3])
screen.cell = as.character(miaccs["assay.cellline",3])
screen.drug = as.character(miaccs["assay.treated",3])
screen.control = as.character(miaccs["assay.untreated",3])
screen.targets = as.character(miaccs["assay.genes",3])
screen.plasmids = as.character(miaccs["assay.plasmids",3])
screen.library = as.character(miaccs["assay.library",3])
screen.NGS = as.character(miaccs["lib.sequencing",3])
screen.despergene = as.character(miaccs["assay.designs",3])
#### FILES AND DATASETS
## PATHs
# abolsute path to where CRISPR-extract.pl and CRISPR-mapping.pl are located
scriptpath=as.character(miaccs["carpools.scriptpath",3])
# abolsute path to where fastq / read count and mapping files are located
datapath = as.character(miaccs["carpools.datapath",3])
## File Names
# name of fastq files of the datasets WITHOUT extension .fastq
# in general, the presence of two replicates for each group is MANDATORY.
# If mapping = TRUE and extract = FALSE, these files need to be already extracted fastq files.
# If mapping = FALSE and extract = FALSE, these files need to be final .txt files with read count data and .txt extension.
fileCONTROL1 = as.character(miaccs["carpools.untreated1",3])
d.CONTROL1 = as.character(miaccs["carpools.untreated1.desc",3])
fileCONTROL2 = as.character(miaccs["carpools.untreated2",3])
d.CONTROL2 = as.character(miaccs["carpools.untreated2.desc",3])
fileTREAT1 = as.character(miaccs["carpools.treated1",3])
d.TREAT1 = as.character(miaccs["carpools.treated1.desc",3])
fileTREAT2 = as.character(miaccs["carpools.treated2",3])
d.TREAT2 = as.character(miaccs["carpools.treated2.desc",3])
# name of reference library .fasta file WITHOUT extension .fasta
referencefile = as.character(miaccs["carpools.lib.ref",3])
## Analysis Name
# This name will be used as file name for all tabular, PDF and HTML files.
analysis.name = stringi::stri_replace_all_fixed(miaccs["assay.name",3], " ", "_")
knitr::opts_chunk$set(fig.path=paste(datapath,"/",analysis.name, "/", sep=""))
#### PERL REGULAR EXPRESSIONS
## Gene Identifier Discrimination from sgRNA name
# Regular expression that is returning the gene identifier from the sgRNA name
# e.g. expression("^(.+?)(_.+)")
g.extractpattern = as.character(miaccs["carpools.regex.gene",3])
# Pattern to extract data if extract=TRUE, for more information see manual
seq.pattern = as.character(miaccs["carpools.regex.extract",3])
# Pattern of READ ID (depends on sequencing maschine), e.g. M01100
maschine.pattern = as.character(miaccs["carpools.regex.maschine",3])
#### DATA EXTRACTION AND MAPPING
## Extract fastq files from fastq raw data?
extract = as.logical(miaccs["carpools.extract.fastq",3])
# is the data within the fastq file in reverse complement oriantation (e.g. paired end read)
reversecomplement = as.logical(miaccs["carpools.fastq.rev",3])
## Mapping
# Shall the extracted fastq file be mapped to the reference library file using the generated or already present .SAM file and count files created?
mapping = as.logical(miaccs["carpools.mapping",3])
## Mapping extracted data
# Create a bowtie2 index file? TRUE if NO bowtie2 index file has been generated for the reference library fasta file, so that a new one is created
createindex = as.logical(miaccs["carpools.bt2.index",3])
# How many threads shall be used for bowtie2 usage?
threads = as.integer(miaccs["carpools.bt2.threads",3])
# Additional parameters for bowtie2 alignment
bowtieparams = as.character(miaccs["carpools.bt2.additional",3])
if(is.na(bowtieparams) || bowtieparams == "NA" || bowtieparams == "none" || is.null(bowtieparams))
{bowtieparams = ""}
# Sensitivity string for bowtie2 alignment, e.g. --very-sensitive-local, --local, --very-sensitive -> have a look in the bowtie2 manual!
sensitivity = as.character(miaccs["carpools.bt2.sens",3])
# How good must the alignment of the 20 nt CRISPR Oligo must be in order to be considered as present in the data?
match = as.character(miaccs["carpools.align.quality",3])
#### REPORT GENERATION
## CONVERT GENE IDENTIFIER?
g.convert = as.logical(miaccs["carpools.gconvert",3])
# Which dataset shall be used?
# See biomaRt for more information on dataset and type
a.database = as.character(miaccs["carpools.bm.database",3])
a.dataset = as.character(miaccs["carpools.bm.dataset",3])
# What kind of identifier is used? e.g. EnsemblID or HGNC Symbol?
# See biomaRt Filter for additional identifiers (filters)
g.identifier = as.character(miaccs["carpools.bm.identifier",3])
# Convert it to?
# see biomaRt for additional attributes
g.identifier.new = as.character(miaccs["carpools.bm.identifier.new",3])
## Anootate final hit candidates?
a.annotate.hits = as.logical(miaccs["carpools.annotate",3])
a.annotate = unlist(strsplit(stringi::stri_replace_all_fixed(miaccs["carpools.bm.attributes",3]," ","", fixed=TRUE), ",", fixed=TRUE))
#### GENERAL
# Add statistics to plots?
plot.statistic = as.logical(miaccs["carpools.statistic",3])
# Column in which the gene/sgRNA identifier is
namecolumn = as.numeric(miaccs["carpools.namecolumn",3])
# Column in which the read count data is
fullmatchcolumn = as.numeric(miaccs["carpools.readcolumn",3])
# Normalize read count of datasets in the plots?
normalize = as.logical(miaccs["carpools.normalize",3])
# Normalization function used to normalize datasets
norm.function = eval(parse(text = miaccs["carpools.normfun",3]))
# color to plot highlighted genes / sgRNAs
plot.labelcolor = as.character(miaccs["carpools.labelcolor",3])
put.names = as.logical(miaccs["carpools.labelnames",3])
#### DATA ANALYSIS
# Targeting Controls
controls.target = as.character(unlist(strsplit(stringi::stri_replace_all_fixed(miaccs["carpools.pos",3]," ", "", fixed=TRUE), ",", fixed=TRUE)))
if(is.na(controls.target) || controls.target == "NULL" || controls.target == "none")
{
controls.target = NULL
}
# Non-Targeting Controls
controls.nontarget = as.character(unlist(strsplit(stringi::stri_replace_all_fixed(miaccs["carpools.non-targeting",3]," ", "", fixed=TRUE), ",", fixed=TRUE)))
if(is.na(controls.nontarget) || controls.nontarget == "NULL" || controls.nontarget == "none" || controls.nontarget == "")
{
controls.nontarget = NULL
}
## wilcoxn
control.picks = as.numeric(miaccs["carpools.randompicks",3])
## CUT OFF
sig.pval.deseq = as.numeric(miaccs["carpools.deseq2",3])
sig.pval.mageck = as.numeric(miaccs["carpools.mageck",3])
sig.pval.wilcox = as.numeric(miaccs["carpools.wilcox",3])
## HIT VISUALIZATION
# This will plot FOUR scatter plots for each enriched or depleted hit automatically and label the genes and sgRNAs
sgRNAs.top.deseq = as.numeric(miaccs["carpools.topdeseq",3])
sgRNAs.top.mageck = as.numeric(miaccs["carpools.topmageck",3])
cutoff.override = as.logical(miaccs["carpools.override",3])
# Percentage of TOP enriched or depleted genes to plot in general
number.hits.plot.enriched = as.numeric(miaccs["carpools.top.enriched",3])
number.hits.plot.depleted = as.numeric(miaccs["carpools.top.depleted",3])
# Comparing Methods
compare.cutoff = as.numeric(miaccs["carpools.overrride.genes",3])
```
\newpage
# Screen
Screen | Information
------ | --------
__Screen ID__ | `r screen.id`
__Screening Date__ | `r screen.date`
__Organism__ | `r screen.organism`
__Cell Line__ | `r screen.cell`
__Drug__ | `r screen.drug`
__Control__ | `r screen.control`
__Number of Target__ | `r screen.targets`
__Designs per Gene__ | `r screen.despergene`
__Library__ | `r screen.library`
__Library Reference File__ | `r referencefile`
__NGS__ | `r screen.NGS`
__MIACCS__
The MIACCS-file can be found at
`r paste("_", datapath, "/", "_" , sep="")`
`r paste("__", "MIACCS.xls" ,"__", sep="")`
__Description__
`r screen.description`
```{r plasmid-screen-overview, echo=FALSE}
## Prepare for plotting overviews
if(miaccs["assay.plasmids.image",3] != "none" && !is.na(miaccs["assay.plasmids.image",3]))
{
plasmid.image = strsplit(miaccs["assay.plasmids.image",3], split=",")
for(i in 1:length(unlist(plasmid.image)))
{
cat("\n")
cat("\\newpage","\n")
cat("\n")
cat("#","Plasmid Overview","\n")
cat("![](",paste(datapath, plasmid.image[[1]][i], sep="/"), ")" )
}
}
if(miaccs["cell.setup.image",3] != "none"&& !is.na(miaccs["cell.setup.image",3]))
{
screen.image = strsplit(miaccs["cell.setup.image",3], split=",")
for(i in 1:length(unlist(screen.image)))
{
cat("\n")
cat("\\newpage","\n")
cat("\n")
cat("#","Experimental Setup","\n")
cat("![](",paste(datapath, screen.image[[1]][i], sep="/"),")" )
}
}
```
\newpage
# Load Data
```{r datainput, echo=FALSE, error=FALSE, warning=FALSE, message=FALSE, results='hide'}
# Do we need to map the data?
if(identical(mapping, TRUE))
{
# Extract Data?
fileCONTROL1 = data.extract(scriptpath, datapath, fastqfile=fileCONTROL1, extract, seq.pattern, maschine.pattern, createindex, referencefile, mapping, reversecomplement, threads, bowtieparams, sensitivity, match)
fileCONTROL2 = data.extract(scriptpath, datapath, fastqfile=fileCONTROL2, extract, seq.pattern, maschine.pattern, createindex, referencefile, mapping, reversecomplement, threads, bowtieparams, sensitivity, match)
fileTREAT1 = data.extract(scriptpath, datapath, fastqfile=fileTREAT1, extract, seq.pattern, maschine.pattern, createindex, referencefile, mapping, reversecomplement, threads, bowtieparams, sensitivity, match)
fileTREAT2 = data.extract(scriptpath, datapath, fastqfile=fileTREAT2, extract, seq.pattern, maschine.pattern, createindex, referencefile, mapping, reversecomplement, threads, bowtieparams, sensitivity, match)
# Data will be mapped, so we start mapping and then get back the data!
CONTROL1 = load.file(paste(datapath, fileCONTROL1, sep="/"))
CONTROL2 = load.file(paste(datapath, fileCONTROL2, sep="/"))
TREAT1 = load.file(paste(datapath, fileTREAT1, sep="/"))
TREAT2 = load.file(paste(datapath, fileTREAT2, sep="/"))
libFILE = load.file(paste(datapath, paste(referencefile,".fasta",sep=""), sep="/"),header = FALSE, type="fastalib")
} else {
# Extracted and mapped read count data is present, so ONLY load files already there using the filename fileCONTROL etc.
CONTROL1 = load.file(paste(datapath, fileCONTROL1, sep="/"))
CONTROL2 = load.file(paste(datapath, fileCONTROL2, sep="/"))
TREAT1 = load.file(paste(datapath, fileTREAT1, sep="/"))
TREAT2 = load.file(paste(datapath, fileTREAT2, sep="/"))
libFILE = load.file(paste(datapath, paste(referencefile,".fasta",sep=""), sep="/"),header = FALSE, type="fastalib")
}
```
__A brief look into the data passed on to the analysis scripts:__
`r # make randoms for brief stats table`
`r sample.sgrna = sample(1:nrow(CONTROL1),10)`
__`r d.CONTROL1`__
__`r fileCONTROL1`__
`r knitr::kable(CONTROL1[sample.sgrna,])`
__`r d.CONTROL2`__
__`r fileCONTROL2`__
`r knitr::kable(CONTROL2[sample.sgrna,])`
__`r d.TREAT1`__
__`r fileTREAT1`__
`r knitr::kable(TREAT1[sample.sgrna,])`
__`r d.TREAT2`__
__`r fileTREAT2`__
`r knitr::kable(TREAT2[sample.sgrna,])`
```{r geneidentifier, echo=FALSE, echo=FALSE, error=FALSE, warning=FALSE, message=FALSE, results='hide'}
if(identical(g.convert, TRUE))
{
cat("## Convert Gene Identifiers", "\n")
cat("Gene identifiers are converted from","\n", "__", g.identifier, "__" ,"to __",g.identifier.new,"__", "using the __", a.dataset,"__ database." ,"\n")
# convert from ENSEMBL ID to human genome nomenclature symbol
CONTROL1 = get.gene.info(CONTROL1, namecolumn=namecolumn, extractpattern=g.extractpattern, database=a.database, dataset=a.dataset, filters=g.identifier, attributes = c(g.identifier.new), return.val = "dataset")
CONTROL2 = get.gene.info(CONTROL2, namecolumn=namecolumn, extractpattern=g.extractpattern, database=a.database, dataset=a.dataset, filters=g.identifier, attributes = c(g.identifier.new), return.val = "dataset")
TREAT1 = get.gene.info(TREAT1, namecolumn=namecolumn, extractpattern=g.extractpattern, database=a.database, dataset=a.dataset, filters=g.identifier, attributes = c(g.identifier.new), return.val = "dataset")
TREAT2 = get.gene.info(TREAT2, namecolumn=namecolumn, extractpattern=g.extractpattern, database=a.database, dataset=a.dataset, filters=g.identifier, attributes = c(g.identifier.new), return.val = "dataset")
# convert control list
controls.target = get.gene.info(as.data.frame(controls.target), namecolumn=namecolumn, extractpattern=g.extractpattern, database=a.database, dataset=a.dataset, filters=g.identifier, attributes = c(g.identifier.new), return.val = "dataset", controls=TRUE)
controls.target = controls.target[,1]
# Library file
libFILE = get.gene.info(libFILE, namecolumn=namecolumn, extractpattern=g.extractpattern, database=a.database, dataset=a.dataset, filters=g.identifier, attributes = c(g.identifier.new), return.val = "dataset")
}
# aggregate to genes for later plots
CONTROL1.g=aggregatetogenes(CONTROL1, agg.function=sum, extractpattern = g.extractpattern)
CONTROL2.g=aggregatetogenes(CONTROL2, agg.function=sum, extractpattern = g.extractpattern)
TREAT1.g=aggregatetogenes(TREAT1, agg.function=sum, extractpattern = g.extractpattern)
TREAT2.g=aggregatetogenes(TREAT2, agg.function=sum, extractpattern = g.extractpattern)
# Add genes names to normal count file (if not present)
CONTROL1=aggregatetogenes(CONTROL1, extractpattern = g.extractpattern, type="annotate")
CONTROL2=aggregatetogenes(CONTROL2, extractpattern = g.extractpattern, type="annotate")
TREAT1=aggregatetogenes(TREAT1, extractpattern = g.extractpattern,type="annotate")
TREAT2=aggregatetogenes(TREAT2, extractpattern = g.extractpattern, type="annotate")
# Calculate Percentage into gene rows
number.enriched.counter = ceiling((nrow(CONTROL1.g)/100)*number.hits.plot.enriched)
number.depleted.counter = ceiling((nrow(CONTROL1.g)/100)*number.hits.plot.depleted)
```
\newpage
# Stats
All file-based output (e.g. tables) is stored in:
_`r datapath`_.
## General
General stats can be found in
__`r paste(analysis.name, "STATS.xls", sep="-")`__.
The following read count statistics were calculated for the single datasets.
```{r stats-general, echo=FALSE}
# General
U1.stats = stats.data(dataset=CONTROL1, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="stats")
U2.stats = stats.data(dataset=CONTROL2, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="stats")
T1.stats = stats.data(dataset=TREAT1, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="stats")
T2.stats = stats.data(dataset=TREAT2, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="stats")
combined.stats = cbind.data.frame(U1.stats[,1:2], U2.stats[,2], T1.stats[,2], T2.stats[,2])
colnames(combined.stats) = c("Readcount", d.CONTROL1, d.CONTROL2, d.TREAT1, d.TREAT2)
# output to report
knitr::kable(combined.stats)
# output to file
xlsx::write.xlsx(combined.stats, file=paste(datapath, paste(analysis.name, "STATS.xls", sep="_"), sep="/"), sheetName="Combined Stats", row.names=FALSE)
```
\newpage
```{r stats-dropout, echo=FALSE}
# General
U1.unmapped = unmapped.genes(data=CONTROL1, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=NULL, extractpattern=g.extractpattern)
U1.unmapped = U1.unmapped[order(U1.unmapped$sgRNA, decreasing=TRUE),]
U2.unmapped = unmapped.genes(data=CONTROL2, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=NULL, extractpattern=g.extractpattern)
U2.unmapped = U2.unmapped[order(U2.unmapped$sgRNA, decreasing=TRUE),]
T1.unmapped = unmapped.genes(data=TREAT1, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=NULL, extractpattern=g.extractpattern)
T1.unmapped = T1.unmapped[order(T1.unmapped$sgRNA, decreasing=TRUE),]
T2.unmapped = unmapped.genes(data=TREAT2, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=NULL, extractpattern=g.extractpattern)
T2.unmapped = T2.unmapped[order(T2.unmapped$sgRNA, decreasing=TRUE),]
# output to file
xlsx::write.xlsx(U1.unmapped, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=d.CONTROL1, row.names=FALSE)
xlsx::write.xlsx(U2.unmapped, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=d.CONTROL2, append=TRUE, row.names=FALSE)
xlsx::write.xlsx(T1.unmapped, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=d.TREAT1, append=TRUE, row.names=FALSE)
xlsx::write.xlsx(T2.unmapped, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=d.TREAT2, append=TRUE, row.names=FALSE)
# Output single sgRNAs of each gene
U1.unmapped.singles = unmapped.genes(data=CONTROL1, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=U1.unmapped[,"name"], extractpattern=g.extractpattern)
U1.unmapped.singles = U1.unmapped.singles[order(U1.unmapped.singles$sgRNA, decreasing=TRUE),]
xlsx::write.xlsx(U1.unmapped.singles, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=paste(d.CONTROL1,"sgRNA"), row.names=FALSE, append=TRUE)
U2.unmapped.singles = unmapped.genes(data=CONTROL2, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=U2.unmapped[,"name"], extractpattern=g.extractpattern)
U2.unmapped.singles = U2.unmapped.singles[order(U2.unmapped.singles$sgRNA, decreasing=TRUE),]
xlsx::write.xlsx(U2.unmapped.singles, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=paste(d.CONTROL2,"sgRNA"), row.names=FALSE, append=TRUE)
T1.unmapped.singles = unmapped.genes(data=TREAT1, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=T1.unmapped[,"name"], extractpattern=g.extractpattern)
T1.unmapped.singles = T1.unmapped.singles[order(T1.unmapped.singles$sgRNA, decreasing=TRUE),]
xlsx::write.xlsx(T1.unmapped.singles, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=paste(d.TREAT1,"sgRNA"), row.names=FALSE, append=TRUE)
T2.unmapped.singles = unmapped.genes(data=TREAT2, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, genes=T2.unmapped[,"name"], extractpattern=g.extractpattern)
T2.unmapped.singles = T2.unmapped.singles[order(T2.unmapped.singles$sgRNA, decreasing=TRUE),]
xlsx::write.xlsx(T2.unmapped.singles, file=paste(datapath, paste(analysis.name, "DROPOUT.xls", sep="_"), sep="/"), sheetName=paste(d.TREAT2,"sgRNA"), row.names=FALSE, append=TRUE)
```
## Missing sgRNAs in Datasets
Information of how many sgRNA per gene were not present in the mapped datasets is stored in:
`r paste("_", datapath, "/", "_", sep="")`
`r paste("__", paste(analysis.name, "DROPOUT.xls", sep="_") ,"__", sep="") `
In brief, the following number of sgRNAs had a _read count of 0_ in the dataset:
__`r d.CONTROL1`:__ __`r sum(U1.unmapped[,2])`__ missing sgRNAs
__`r d.CONTROL2`:__ __`r sum(U2.unmapped[,2])`__ missing sgRNAs
__`r d.TREAT1`:__ __`r sum(T1.unmapped[,2])`__ missing sgRNAs
__`r d.TREAT2`:__ __`r sum(T2.unmapped[,2])`__ missing sgRNAs
## All Stats
In depth dataset read count stats can be found in
`r paste("_", datapath, "/", "_" , sep="")`
`r paste("__", paste(analysis.name, "STATS.xls", sep="-") ,"__", sep="" ) `
```{r, stats-all, echo=FALSE}
U1.allstats = stats.data(dataset=CONTROL1, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="dataset")
U2.allstats = stats.data(dataset=CONTROL2, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="dataset")
T1.allstats = stats.data(dataset=TREAT1, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="dataset")
T2.allstats = stats.data(dataset=TREAT2, namecolumn = namecolumn, fullmatchcolumn = fullmatchcolumn, extractpattern=g.extractpattern, type="dataset")
combined.stats.mean = cbind.data.frame(U1.allstats[,"Name"], U1.allstats[,"readcount.mean"],U2.allstats[,"readcount.mean"],T1.allstats[,"readcount.mean"],T2.allstats[,"readcount.mean"])
colnames(combined.stats.mean) = c("Gene",d.CONTROL1, d.CONTROL2, d.TREAT1, d.TREAT2)
combined.stats.median = cbind.data.frame(U1.allstats[,"Name"], U1.allstats[,"readcount.median"],U2.allstats[,"readcount.median"],T1.allstats[,"readcount.median"],T2.allstats[,"readcount.median"])
colnames(combined.stats.median) = c("Gene",d.CONTROL1, d.CONTROL2, d.TREAT1, d.TREAT2)
combined.stats.min = cbind.data.frame(U1.allstats[,"Name"], U1.allstats[,"readcount.min"],U2.allstats[,"readcount.min"],T1.allstats[,"readcount.min"],T2.allstats[,"readcount.min"])
colnames(combined.stats.min) = c("Gene",d.CONTROL1, d.CONTROL2, d.TREAT1, d.TREAT2)
combined.stats.max = cbind.data.frame(U1.allstats[,"Name"], U1.allstats[,"readcount.max"],U2.allstats[,"readcount.max"],T1.allstats[,"readcount.max"],T2.allstats[,"readcount.max"])
colnames(combined.stats.max) = c("Gene",d.CONTROL1, d.CONTROL2, d.TREAT1, d.TREAT2)
xlsx::write.xlsx(combined.stats.mean, file=paste(datapath, paste(analysis.name, "STATS.xls", sep="_"), sep="/"), sheetName="Mean Read Count", append=TRUE ,row.names=FALSE)
xlsx::write.xlsx(combined.stats.median, file=paste(datapath, paste(analysis.name, "STATS.xls", sep="_"), sep="/"), sheetName="Median Read Count", append=TRUE, row.names=FALSE)
xlsx::write.xlsx(combined.stats.min, file=paste(datapath, paste(analysis.name, "STATS.xls", sep="_"), sep="/"), sheetName="Minimum Read Count", append=TRUE, row.names=FALSE)
xlsx::write.xlsx(combined.stats.max, file=paste(datapath, paste(analysis.name, "STATS.xls", sep="_"), sep="/"), sheetName="Maximum Read Count", append=TRUE, row.names=FALSE)
```
\newpage
# Quality Control
## Read Distribution
These plots show how the read count of the sgRNAs for each dataset is distributed.
Depending on the treatment stringency, e.g. in resistance or dropout screens, the data can show asymmetry.
However, the major population should be more or less normally distributed.
```{r QC-distribution, echo=FALSE, sanitize=TRUE}
par(mfrow=c(1,1))
carpools.read.distribution(CONTROL1, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.CONTROL1, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic)
carpools.read.distribution(CONTROL1, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.CONTROL1, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic, type="whisker")
carpools.read.distribution(CONTROL2, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.CONTROL2, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic)
carpools.read.distribution(CONTROL2, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.CONTROL2, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic, type="whisker")
# add new page
cat("\\newpage", "\n")
carpools.read.distribution(TREAT1, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.TREAT1, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic)
carpools.read.distribution(TREAT1, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.TREAT1, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic, type="whisker")
carpools.read.distribution(TREAT2, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.TREAT2, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic)
carpools.read.distribution(TREAT2, fullmatchcolumn=fullmatchcolumn,breaks=200, title=d.TREAT2, xlab="log2 Readcount", ylab="# sgRNAs",statistics=plot.statistic, type="whisker")
```
\pagebreak
## Read Depth
The following plot shows the read count for each gene normalized to the number of sgRNAs. Spikes indicate a higher read count per sgRNA for this particular gene.
One would expect no outstanding spikes within the untreated data samples, however spikes within the treated datasets indicate a read count enrichment for this particular gene.
If a non-targeting control has been set in the MIACCS file, this control is highlighted in __orange__ color.
```{r QC-readdepth, echo=FALSE, sanitize=TRUE}
#, fig.width=10, fig.height=20
#par(mfrow=c(4,1))
carpools.read.depth(datasets = list(CONTROL1,CONTROL2,TREAT1,TREAT2), namecolumn=namecolumn ,fullmatchcolumn=fullmatchcolumn, dataset.names=list(d.CONTROL1,d.CONTROL2,d.TREAT1,d.TREAT2), extractpattern=g.extractpattern, xlab="Genes", ylab="Read Count per sgRNA",statistics=plot.statistic, labelgenes = NULL, controls.target = controls.target, controls.nontarget=controls.nontarget)
par(mfrow=c(1,1))
```
\pagebreak
## Designs per Gene
These plots provide an overview of the representation of sgRNAs per gene within your data.
Depending on the number of sgRNAs per gene in the library, one would expect a representation of more than 80 % of sgRNAs per gene in the untreated samples.
Moreover, genes with a low percentage of present sgRNAs will also show a reduced readcount.
```{r QC-despergene, echo=FALSE, sanitize=TRUE}
#, fig.width=10, fig.height=10
par(mfrow=c(2,2))
control1.readspergene = carpools.reads.genedesigns(CONTROL1, fullmatchcolumn=fullmatchcolumn, namecolumn=namecolumn, title=paste("% sgRNAs:", d.CONTROL1, sep=" "), xlab="% of sgRNAs present", ylab="# of Genes")
control2.readspergene = carpools.reads.genedesigns(CONTROL2, fullmatchcolumn=fullmatchcolumn, namecolumn=namecolumn, title=paste("% sgRNAs:", d.CONTROL2, sep=" "), xlab="% of sgRNAs present", ylab="# of Genes")
treat1.readspergene = carpools.reads.genedesigns(TREAT1, fullmatchcolumn=fullmatchcolumn, namecolumn=namecolumn, title=paste("% sgRNAs:", d.TREAT1, sep=" "), xlab="% of sgRNAs present", ylab="# of Genes")
treat2.readspergene = carpools.reads.genedesigns(TREAT2, fullmatchcolumn=fullmatchcolumn, namecolumn=namecolumn, title=paste("% sgRNAs:", d.TREAT2, sep=" "), xlab="% of sgRNAs present", ylab="# of Genes")
par(mfrow=c(1,1))
```
\newpage
```{r echo=FALSE, eval=FALSE}
if(!is.null(controls.nontarget))
{
cat("\n")
#cat("\\blandscape","\n")
}
```
## Controls
### Non-Targeting
Non-targeting controls are sgRNAs that do either not target the genome at all (so called random or scramble sgRNAs) or target a gene that does not show a phenotype in the screen.
Therefore, the scatter for these, which are highlighted in __blue__, will be distributed within the main cloud of scatter points.
`r if(is.null(controls.nontarget)) {print("Unfortunately, no Non-Targeting controls were set.")}`
```{r QC-non-targeting-scatter, echo=FALSE, sanitize=TRUE, fig.width=8, fig.height=8}
# Plot non-targeting controls
#, fig.height=20, fig.width=30,
if(!is.null(controls.nontarget))
{
cat("Non-targeting control: ", controls.nontarget ,"\n")
cat("\n")
carpools.read.count.vs(dataset=list(TREAT1, TREAT2, CONTROL1, CONTROL2), dataset.names = c(d.TREAT1, d.TREAT2, d.CONTROL1, d.CONTROL2), pairs=TRUE, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, title=analysis.name, pch=16, normalize=normalize, norm.function=norm.function, labelgenes=controls.nontarget, labelcolor="blue", center=FALSE, aggregated=FALSE)
} else {
cat("No non-targeting control has been set in the MIACCS file.")
}
```
```{r echo=FALSE, eval=FALSE}
if(!is.null(controls.nontarget))
{
cat("\n")
#cat("\\elandscape","\n")
}
```
\newpage
### Positive Controls
SgRNAs targeting genes that will show a phenotype in the screening setup can be used as positive controls.
These will show either an enrichment (in resistance screens) or a depletion (in dropout screens) in the treatment.
Within the scatter
```{r QC-targeting-scatter, echo=FALSE, sanitize=TRUE, fig.width=8, fig.height=8}
# Plot non-targeting controls
#, fig.height=20, fig.width=30
if(!is.null(controls.target))
{
cat("Positive Control: ", as.character(controls.target) ,"\n")
cat("\n")
carpools.read.count.vs(dataset=list(TREAT1.g, TREAT2.g, CONTROL1.g, CONTROL2.g), dataset.names = c(d.TREAT1, d.TREAT2, d.CONTROL1, d.CONTROL2), pairs=TRUE, namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, title=analysis.name, pch=16, normalize=normalize, norm.function=norm.function, labelgenes=controls.target, labelcolor="red", center=FALSE, aggregated=TRUE)
} else {
cat("No positive control has been set in the MIACCS file.")
}
```
\newpage
# Hit Analysis
Hit analsysis is performed using three different methods:
* Wilcox
* DESeq2
* MAGeCK
For each analysis method, separate plots will be created and analysis files will be written to __`r paste(analysis.name, "HIT-CALLING.xls", sep="_")`__. See below for further information.
The following adjusted p-values are used to determine significance levels:
Method | p-value
------ | -----
Wilcox | `r sig.pval.wilcox`
DESeq2 | `r sig.pval.deseq`
MAGeCK | `r sig.pval.mageck`
__Wilcox__
Within this approach, the read counts of all sgRNAs in one dataset are first normalized by the function set in the MIACCS file. By default, normalization is done by read count division with the dataset median.
Then, the fold change of each population of sgRNAs for a gene is tested against the population of either the non-targeting controls or randomly picked sgRNAs, as defined by the random picks option within the MIACCS file, using a two-sided Mann-Whitney test with FDR correction.
__DESeq2__
For the DESeq2 analysis implementation, the read counts of all sgRNAs for a given gene are first summed up to increase the available read count.
Then, DESeq2 analysis is perfomed, which includes the estimation of size-factors, the variance stabilization using a parametric fit and a Wald-Test for differnece in log2 fold changes between the untreated and treated data.
More information about this can be found in _Love et al._
[Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2](http://www.ncbi.nlm.nih.gov/pubmed/25516281)
_Genome Biology_ 2014
__MAGeCK__
MAGeCK analysis uses a rank-based model to test for a change in abundance of sgRNAs after median normalization of the dataset.
Further information can be found at the [MAGeCK Homepage](http://sourceforge.net/projects/mageck/).
\newpage
## Wilcox
All analysis data can be found in the __`r paste(analysis.name, "HIT-CALLING.xls", sep="_")`__ file.
```{r HT-method-wilcox, echo=FALSE}
# wilcox
if(is.null(controls.nontarget))
{
cat("\n")
cat(paste("**Since no non-targeting controls were set,", control.picks, "sgRNAs are picked from the datset as reference population.**", sep=" "))
cat("\n")
}
data.wilcox = stat.wilcox(untreated.list = list(CONTROL1, CONTROL2), treated.list = list(TREAT1,TREAT2), namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, normalize=normalize, norm.fun=norm.function, sorting=FALSE, controls=controls.nontarget, control.picks=control.picks)
xlsx::write.xlsx(data.wilcox, file=paste(datapath, paste(analysis.name, "HIT-CALLING.xls", sep="_"), sep="/"), sheetName="Wilcox")
```
### P-Value Distribution
For all genes present in the data, the -log10 corrected p-values are plotted to estimate how well the analysis method performed for this screen. A straight line of points with only small differences in the p-value indicates that the analysis method did not perform well.
Genes that resulted in a p-value below the threshold set in the MIACCS file are highlighted in red color.
```{r HT-wilcox-pval-distribution, echo=FALSE}
carpools.waterfall.pval(type="wilcox", dataset=data.wilcox, pval=sig.pval.wilcox, log=TRUE)
```
### Enriched
The following genes showed enrichment in the treatment datasets with a __p-value smaller than `r sig.pval.wilcox`__:
```{r HT-method-wilcox-enriched, echo=FALSE}
data.wilcox.plot.enriched = data.wilcox[data.wilcox$foldchange > 1,]
data.wilcox.plot.enriched = data.wilcox.plot.enriched[order(data.wilcox.plot.enriched$p.value, na.last=TRUE),]
if(nrow(data.wilcox.plot.enriched[data.wilcox.plot.enriched$p.value < sig.pval.wilcox,]) > 0)
{
knitr::kable(data.wilcox.plot.enriched[data.wilcox.plot.enriched$p.value < sig.pval.wilcox,])
} else { cat("**No genes showed significant enrichment with a p-value lower than", sig.pval.wilcox, "**")}
```
According to the value set, the __top `r number.hits.plot.enriched` % enriched genes__ were:
`r knitr::kable(data.wilcox.plot.enriched[1:number.enriched.counter,])`
\newpage
### Depleted
The following genes showed depletion in the treatment datasets with a __p-value smaller than `r sig.pval.wilcox`__:
```{r HT-method-wilcox-depleted, echo=FALSE}
data.wilcox.plot.depleted = data.wilcox[data.wilcox$foldchange < 1,]
data.wilcox.plot.depleted = data.wilcox.plot.depleted[order(data.wilcox.plot.depleted$p.value, na.last=TRUE),]
if(nrow(data.wilcox.plot.depleted[data.wilcox.plot.depleted$p.value < sig.pval.wilcox,]) > 0)
{
knitr::kable(data.wilcox.plot.depleted[data.wilcox.plot.depleted$p.value < sig.pval.wilcox,])
} else { cat("**No genes showed significant depletion with a p-value lower than", sig.pval.wilcox, "**")}
```
According to the value set, the __top `r number.hits.plot.depleted` % depleted genes__ were:
`r knitr::kable(data.wilcox.plot.depleted[1:number.depleted.counter,])`
\newpage
## DESeq2
All analysis data can be found in the __`r paste(analysis.name, "HIT-CALLING.xls", sep="_")`__ file.
```{r HT-method-DESeq2, echo=FALSE, message=FALSE,warning=FALSE}
# DESeq2
data.deseq = stat.DESeq(untreated.list = list(CONTROL1, CONTROL2), treated.list = list(TREAT1,TREAT2), namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, extractpattern=g.extractpattern, sorting=FALSE, filename.deseq = paste(analysis.name, "-ANALYSIS-DESeq2-sgRNA.tab", sep=""), sgRNA.pval = sig.pval.deseq, fitType="mean")
xlsx::write.xlsx(as.data.frame(data.deseq$genes), file=paste(datapath, paste(analysis.name, "HIT-CALLING.xls", sep="_"), sep="/"), sheetName="DESeq2", append=TRUE)
```
### P-Value Distribution
For all genes present in the data, the -log10 corrected p-values are plotted to estimate how well the analysis method performed for this screen. A straight line of points with only small differences in the p-value indicates that the analysis method did not perform well.
Genes that resulted in a p-value below the threshold set in the MIACCS file are highlighted in red color.
```{r HT-deseq-pval-distribution, echo=FALSE}
carpools.waterfall.pval(type="deseq2", dataset=data.deseq, pval=sig.pval.wilcox, log=TRUE)
```
### Enriched
The following genes showed enrichment in the treatment datasets with a __p-value smaller than `r sig.pval.deseq`__:
```{r HT-method-deseq-enriched, echo=FALSE}
data.deseq.plot.enriched = as.data.frame(data.deseq[[1]][data.deseq[[1]]$log2FoldChange > 0,])
data.deseq.plot.enriched = data.deseq.plot.enriched[order(data.deseq.plot.enriched$padj, na.last=TRUE),]
if(nrow(data.deseq.plot.enriched[data.deseq.plot.enriched$padj < sig.pval.deseq,c(2,3,6,8)]) > 0)
{
knitr::kable(data.deseq.plot.enriched[data.deseq.plot.enriched$padj < sig.pval.deseq,c(2,3,6,8)])
} else { cat("**No genes showed significant enrichment with a p-value lower than", sig.pval.deseq, "**")}
```
According to the value set, the __top `r number.hits.plot.enriched` % enriched genes__ were:
`r knitr::kable(data.deseq.plot.enriched[1:number.enriched.counter,c(2,3,6,8)])`
\newpage
### Depleted
The following genes showed depletion in the treatment datasets with a __p-value smaller than `r sig.pval.deseq`__:
```{r HT-method-deseq-depleted, echo=FALSE}
data.deseq.plot.depleted = data.deseq[[1]][data.deseq[[1]]$log2FoldChange < 0,]
data.deseq.plot.depleted = data.deseq.plot.depleted[order(data.deseq.plot.depleted$padj, na.last=TRUE),]
if(nrow(data.deseq.plot.depleted[data.deseq.plot.depleted$padj < sig.pval.deseq,c(2,3,6,8)]) > 0)
{
knitr::kable(data.deseq.plot.depleted[data.deseq.plot.depleted$padj < sig.pval.deseq,c(2,3,6,8)])
} else { cat("**No genes showed significant depletion with a p-value lower than", sig.pval.deseq, "**")}
```
According to the value set, the __top `r number.hits.plot.depleted` % depleted genes__ were:
`r knitr::kable(data.deseq.plot.depleted[1:number.depleted.counter,c(2,3,6,8)])`
\newpage
## MAGeCK
All analysis data for MAGeCK can be found in the __`r paste(analysis.name, "HIT-CALLING.xls", sep="_")`__ file.
```{r HT-method-MAGeCK, echo=FALSE}
# MAGeCK
data.mageck = stat.mageck(untreated.list = list(CONTROL1, CONTROL2), treated.list = list(TREAT1,TREAT2), namecolumn=namecolumn, fullmatchcolumn=fullmatchcolumn, norm.fun="median", extractpattern=g.extractpattern, mageckfolder=NULL, sort.criteria="neg", adjust.method="fdr", filename = paste(analysis.name, "-ANALYSIS-MAGeCK-RAW", sep=""), fdr.pval=sig.pval.mageck)
xlsx::write.xlsx(as.data.frame(data.mageck$genes), file=paste(datapath, paste(analysis.name, "HIT-CALLING.xls", sep="_"), sep="/"), sheetName="MAGeCK", append=TRUE)
# get number of sig gene
mageck.cutoff = nrow(data.mageck[[1]][data.mageck[[1]]$pos < sig.pval.mageck,]) + nrow(data.mageck[[1]][data.mageck[[1]]$neg < sig.pval.mageck,])
```
### P-Value Distribution
For all genes present in the data, the -log10 corrected p-values are plotted to estimate how well the analysis method performed for this screen. A straight line of points with only small differences in the p-value indicates that the analysis method did not perform well.
Genes that resulted in a p-value below the threshold set in the MIACCS file are highlighted in red color.
```{r HT-mageck-pval-distribution, echo=FALSE}
carpools.waterfall.pval(type="mageck", dataset=data.mageck, pval=sig.pval.wilcox, log=TRUE)
```
### Enriched
The following genes showed enrichment in the treatment datasets with __a p-value smaller than `r sig.pval.mageck`__:
```{r HT-method-mageck-enriched, echo=FALSE}
data.mageck.plot.enriched = data.mageck[[1]][data.mageck[[1]]$pos < sig.pval.mageck,]
data.mageck.plot.enriched = data.mageck.plot.enriched[order(data.mageck.plot.enriched$rank.pos, na.last=TRUE),]
if(nrow(data.mageck.plot.enriched) > 0)
{
knitr::kable(data.mageck.plot.enriched[,c(2,3,4,5,6,7)])
} else { cat("**No genes showed significant enrichment with a p-value lower than", sig.pval.mageck, "**")}
data.mageck.plot.enriched = data.mageck[[1]][order(data.mageck[[1]]$rank.pos, na.last=TRUE),]
```
According to the value set, the __top `r number.hits.plot.enriched` % enriched genes__ were:
`r knitr::kable(data.mageck.plot.enriched[1:number.enriched.counter,c(2,3,4,5,6,7)])`
\newpage
### Depleted
The following genes showed depletion in the treatment datasets with a __p-value smaller than `r sig.pval.mageck`__:
```{r HT-method-mageck-depleted, echo=FALSE}
data.mageck.plot.depleted = data.mageck[[1]][data.mageck[[1]]$neg < sig.pval.mageck,]
data.mageck.plot.depleted = data.mageck.plot.depleted[order(data.mageck.plot.depleted$rank.neg, na.last=TRUE),]
if(nrow(data.mageck.plot.depleted) > 0)
{
knitr::kable(data.mageck.plot.depleted[,c(2,3,4,5,6,7)])
} else { cat("**No genes showed significant depletion with a p-value lower than", sig.pval.mageck, "**")}
data.mageck.plot.depleted = data.mageck[[1]][order(data.mageck[[1]]$rank.neg, na.last=TRUE),]
```
According to the value set, the __top `r number.hits.plot.depleted` % depleted genes__ were:
`r knitr::kable(data.mageck.plot.depleted[1:number.depleted.counter,c(2,3,4,5,6,7)])`
\newpage
# Hit Candidate Overview
## Overview
Genes which showed enrichment or depletion within the individual analysis methods are presented in the following section.
All genes that showed _significant_ __enrichment__ within Wilcox, DESeq2 and MAGeCK with the given p-value cutoffs are highlighted in __red__ color.
All genes that showed _significant_ __depletion__ within Wilcox, DESeq2 and MAGeCK with the given p-value cutoffs are highlighted in __blue__ color.
Moreover, all genes that showed up as _significantly_ enriched or depleted in any of the analysis methods are highlighted in __orange__ color.
Genes that showed no significant effect are presented in grey or black color.
```{r HT-overview-hits, echo=FALSE, sanitize=TRUE}
carpools.hit.overview(wilcox=data.wilcox, deseq=data.deseq, mageck=data.mageck, cutoff.deseq = sig.pval.deseq, cutoff.wilcox = sig.pval.wilcox, cutoff.mageck = sig.pval.mageck, cutoff.override=cutoff.override, cutoff.hits=NULL, plot.genes="overlapping")
```
```{r HT-overlaps, echo=FALSE, sanitize=TRUE}
overlap.enriched = generate.hits(wilcox=data.wilcox, deseq=data.deseq, mageck=data.mageck, type="enriched", cutoff.deseq = sig.pval.deseq, cutoff.wilcox = sig.pval.wilcox, cutoff.mageck = sig.pval.mageck, cutoff.override=cutoff.override, cutoff.hits=compare.cutoff, plot.genes="overlapping")
overlap.depleted = generate.hits(wilcox=data.wilcox, deseq=data.deseq, mageck=data.mageck, type="depleted", cutoff.deseq = sig.pval.deseq, cutoff.wilcox = sig.pval.wilcox, cutoff.mageck = sig.pval.mageck, cutoff.override=cutoff.override, cutoff.hits=compare.cutoff, plot.genes="overlapping")
# Combine datasets for overlapping hits
overlap.enriched.dataset = data.frame(
gene = overlap.enriched,
log2FC = data.deseq[[1]][rownames(data.deseq[[1]]) %in% overlap.enriched, "log2FoldChange"],
wilcox.pval = data.wilcox[rownames(data.wilcox) %in% overlap.enriched,"p.value"],
deseq2.pval = data.deseq[[1]][rownames(data.deseq[[1]]) %in% overlap.enriched,"padj"],
mageck.pval = data.mageck[[1]][rownames(data.mageck[[1]]) %in% overlap.enriched,"pos"],
stringsAsFactors=FALSE)
overlap.depleted.dataset = data.frame(
gene = overlap.depleted,
log2FC = data.deseq[[1]][rownames(data.deseq[[1]]) %in% overlap.depleted, "log2FoldChange"],
wilcox.pval = data.wilcox[rownames(data.wilcox) %in% overlap.depleted,"p.value"],
deseq2.pval = data.deseq[[1]][rownames(data.deseq[[1]]) %in% overlap.depleted,"padj"],
mageck.pval = data.mageck[[1]][rownames(data.mageck[[1]]) %in% overlap.depleted,"neg"],
stringsAsFactors=FALSE)
```
\newpage
## Overlaps in Enrichment Analysis
```{r HT-comparevenn-enriched, echo=FALSE, sanitize=TRUE}
if(nrow(overlap.enriched.dataset) > 0)
{
cat("The following genes showed enrichment in all three analysis methods.", "\n")
knitr::kable(overlap.enriched.dataset)
cutoff.override.enriched = cutoff.override
} else {
cat("\n")
cat("**ATTENTION** \n")
cat("**No overlapping enriched genes were found between all three methods.** ", "\n \n")
cat("Therefore, the overlap from the", number.hits.plot.enriched, "% of genes within all methods is used for plotting possible gene candidates.","\n" )
cat("\n")
cat("We strongly advise you to have a closer look at the individual outputs of the analysis methods and to carefully look at the following candidate genes.")
cutoff.override.enriched=TRUE
overlap.enriched = generate.hits(wilcox=data.wilcox, deseq=data.deseq, mageck=data.mageck, type="enriched", cutoff.deseq = sig.pval.deseq, cutoff.wilcox = sig.pval.wilcox, cutoff.mageck = sig.pval.mageck, cutoff.override= cutoff.override.enriched, cutoff.hits=number.enriched.counter, plot.genes="overlapping")
overlap.enriched.dataset = data.frame(
gene = overlap.enriched,
log2FC = data.deseq[[1]][rownames(data.deseq[[1]]) %in% overlap.enriched, "log2FoldChange"],
wilcox.pval = data.wilcox[rownames(data.wilcox) %in% overlap.enriched,"p.value"],
deseq2.pval = data.deseq[[1]][rownames(data.deseq[[1]]) %in% overlap.enriched,"padj"],
mageck.pval = data.mageck[[1]][rownames(data.mageck[[1]]) %in% overlap.enriched,"pos"],
stringsAsFactors=FALSE)
}
```
Within the __top enriched hits__, the overlap of enriched hits per analysis method is displayed as follows:
```{r HT-compare-enriched-overlap, echo=FALSE,sanitize=TRUE, message=FALSE,warning=FALSE}
venn.enriched = compare.analysis(wilcox=data.wilcox, deseq=data.deseq, mageck=data.mageck, type="enriched", cutoff.deseq = sig.pval.deseq, cutoff.wilcox = sig.pval.wilcox, cutoff.mageck = sig.pval.mageck, cutoff.override=cutoff.override.enriched, cutoff.hits=number.enriched.counter, output="venn")
require(VennDiagram)
grid::grid.draw(VennDiagram::venn.diagram(venn.enriched, file=NULL, fill=c("lightgreen","lightblue2","lightgray"), na="remove", cex=2,lty=2, cat.cex=2))
```
\newpage
## Overlaps in Depletion Analysis
```{r HT-comparevenn-depleted, echo=FALSE, sanitize=TRUE}
if(nrow(overlap.depleted.dataset) > 0)
{
cat("The following genes showed depletion in all three analysis methods.", "\n")
knitr::kable(overlap.depleted.dataset)
cutoff.override.depleted=cutoff.override
} else {
cat("\n")
cat("**ATTENTION** \n")
cat("**No overlapping depleted genes were found between all three methods.**", "\n \n")
cat("Therefore, the overlap from the", number.hits.plot.depleted, "% of genes within all methods is used for plotting possible hit candidates.","\n" )
cat("\n")
cat("We strongly advise you to have a closer look at the individual outputs of the analysis methods and to carefully look at the following candidate genes.")
cutoff.override.depleted = TRUE
overlap.depleted = generate.hits(wilcox=data.wilcox, deseq=data.deseq, mageck=data.mageck, type="depleted", cutoff.deseq = sig.pval.deseq, cutoff.wilcox = sig.pval.wilcox, cutoff.mageck = sig.pval.mageck, cutoff.override=cutoff.override.depleted, cutoff.hits=number.depleted.counter, plot.genes="overlapping")