search

SBU-BMI / tumor-til-survival-analysis

Scripts to run analysis of tumor/TILs.
Apache License 2.0
2 stars 0 forks source link

add survival output generation #7

Closed kaczmarj closed 1 year ago

kaczmarj commented 1 year ago

this pr adds the survival component of the analysis pipeline. @lthealy - i had to add many things to get this working properly so i opened this pr. this replaces #6.

i get this error when using the sample survival CSV in the til align repo (happens when rendering the r markdown). do you have a dataset we can use to test the full pipeline with? i'm assuming this happens because there are not enough samples?

we should probably account for this potential error.. depending on why it happens. for instance what if users run into this. what should we tell them to do?

processing file: Descriptive_Statistics.rmd
  |...                                                                   |   5%
   inline R code fragments

  |......                                                                |   9%
label: setup (with options) 
List of 1
 $ include: logi FALSE

  |..........                                                            |  14%
  ordinary text without R code

  |.............                                                         |  18%
label: load (with options) 
List of 1
 $ include: logi FALSE

  |................                                                      |  23%
  ordinary text without R code

  |...................                                                   |  27%
label: unnamed-chunk-1 (with options) 
List of 1
 $ fig.height: num 4

  |......................                                                |  32%
   inline R code fragments

  |.........................                                             |  36%
label: unnamed-chunk-2
Quitting from lines 67-83 (Descriptive_Statistics.rmd) 
Error in chisq.test(chi) : at least one entry of 'x' must be positive
Calls: <Anonymous> ... eval_with_user_handlers -> eval -> eval -> chisq.test
In addition: Warning messages:
1: Removed 4 rows containing non-finite values (stat_bin). 
2: In chisq.test(chi) : Chi-squared approximation may be incorrect
3: In chisq.test(chi) : Chi-squared approximation may be incorrect

related to #3

kaczmarj commented 1 year ago

@lthealy - i still get the error Error in chisq.test(chi) : at least one entry of 'x' must be positive when using the larger survival info csv. can you please take a look? i am using the til and brca tumor outputs that you shared with me on box.

Warning message:
Removed 1 rows containing non-finite values (stat_bin). 
null device 
          1 
Running survival pipeline...
$csvPath
[1] "/data/output.csv"

$survTime
[1] "survivalA"

$survCensor
[1] "censorA.0yes.1no"

$outputFormat
[1] "pdf_document"

processing file: Descriptive_Statistics.rmd
  |...                                                                   |   5%
   inline R code fragments

  |......                                                                |   9%
label: setup (with options) 
List of 1
 $ include: logi FALSE

  |..........                                                            |  14%
  ordinary text without R code

  |.............                                                         |  18%
label: load (with options) 
List of 1
 $ include: logi FALSE

  |................                                                      |  23%
  ordinary text without R code

  |...................                                                   |  27%
label: unnamed-chunk-1 (with options) 
List of 1
 $ fig.height: num 4

  |......................                                                |  32%
   inline R code fragments

  |.........................                                             |  36%
label: unnamed-chunk-2
Quitting from lines 67-83 (Descriptive_Statistics.rmd) 
Error in chisq.test(chi) : at least one entry of 'x' must be positive
Calls: <Anonymous> ... eval_with_user_handlers -> eval -> eval -> chisq.test
In addition: Warning messages:
1: Removed 983 rows containing non-finite values (stat_bin). 
2: In chisq.test(chi) : Chi-squared approximation may be incorrect
3: In chisq.test(chi) : Chi-squared approximation may be incorrect

during the til alignment phase, many of the values that are printed are close to 1.75. is this expected?

[1] "Sample 1: 1.74747474747475"
[1] "Sample 2: 1.74747474747475"
[1] "Sample 3: 1.74747474747475"
[1] "Sample 4: 1.74747474747475"
[1] "Sample 5: 1.74747474747475"
[1] "Sample 6: 1.74747474747475"
[1] "Sample 7: 1.74747474747475"
[1] "Sample 8: 1.74747474747475"
[1] "Sample 9: 1.74747474747475"
[1] "Sample 10: 1.74747474747475"
[1] "Sample 11: 1.74747474747475"
[1] "Sample 12: 1.74747474747475"
[1] "Sample 13: 1.74747474747475"
[1] "Sample 14: 1.74509803921569"
[1] "Sample 15: 1.74509803921569"
[1] "Sample 16: 1.74509803921569"
[1] "Sample 17: 1.74747474747475"
[1] "Sample 18: 1.74747474747475"
[1] "Sample 19: 1.75"
[1] "Sample 20: 1.75"
[1] "Sample 21: 1.75"
[1] "Sample 22: 1.75"
[1] "Sample 23: 1.75"
[1] "Sample 24: 1.75"
[1] "Sample 25: 1.75"
[1] "Sample 26: 1.75"
[1] "Sample 27: 1.75116279069767"
[1] "Sample 28: 1.74509803921569"
[1] "Sample 29: 1.74509803921569"
[1] "Sample 30: 1.74509803921569"
[1] "Sample 31: 1.74509803921569"
[1] "Sample 32: 1.74747474747475"
[1] "Sample 33: 1.76"
[1] "Sample 34: 1.74509803921569"
[1] "Sample 35: 1.74747474747475"
[1] "Sample 36: 1.74747474747475"
[1] "Sample 37: 1.74747474747475"
[1] "Sample 38: 1.74747474747475"
[1] "Sample 39: 1.74747474747475"
[1] "Sample 40: 1.74747474747475"
[1] "Sample 41: 1.74747474747475"
[1] "Sample 42: 1.74747474747475"
[1] "Sample 43: 1.74747474747475"
[1] "Sample 44: 1.74747474747475"
[1] "Sample 45: 1.74509803921569"
[1] "Sample 46: 1.74509803921569"
[1] "Sample 47: 1.74509803921569"
[1] "Sample 48: 1.74509803921569"
[1] "Sample 49: 1.74509803921569"
[1] "Sample 50: 1.74509803921569"
kaczmarj commented 1 year ago

there is a syntax error in Descriptive_Statistics.rmd at https://github.com/SBU-BMI/til_align/commit/d25adf5e778ab9aafc97ef0c1d361d72f5af7daa

Quitting from lines 63-77 (Descriptive_Statistics.rmd) 
Error: Failed to parse the inline R code: sum(missingTIL == "no cancer (Reason: <text>:1:19: unexpected INCOMPLETE_STRING
1: sum(missingTIL == "no cancer
                      ^)
kaczmarj commented 1 year ago

@lthealy I still get an error with chisq.test

processing file: Descriptive_Statistics.rmd
  |...                                                                   |   5%
   inline R code fragments

  |......                                                                |   9%
label: setup (with options) 
List of 1
 $ include: logi FALSE

  |..........                                                            |  14%
  ordinary text without R code

  |.............                                                         |  18%
label: load (with options) 
List of 1
 $ include: logi FALSE

  |................                                                      |  23%
  ordinary text without R code

  |...................                                                   |  27%
label: unnamed-chunk-1 (with options) 
List of 1
 $ fig.height: num 4

  |......................                                                |  32%
   inline R code fragments

  |.........................                                             |  36%
label: unnamed-chunk-2
Quitting from lines 79-107 (Descriptive_Statistics.rmd) 
Error in chisq.test(chi) : 'x' must at least have 2 elements
Calls: <Anonymous> ... eval_with_user_handlers -> eval -> eval -> chisq.test
In addition: Warning message:
Removed 99 rows containing non-finite values (stat_bin). 

Execution halted
kaczmarj commented 1 year ago

closer but still getting an error. now with pandoc:

processing file: Descriptive_Statistics.rmd
  |...                                                                   |   5%
   inline R code fragments

  |......                                                                |   9%
label: setup (with options) 
List of 1
 $ include: logi FALSE

  |..........                                                            |  14%
  ordinary text without R code

  |.............                                                         |  18%
label: load (with options) 
List of 1
 $ include: logi FALSE

  |................                                                      |  23%
  ordinary text without R code

  |...................                                                   |  27%
label: unnamed-chunk-1 (with options) 
List of 1
 $ fig.height: num 4

  |......................                                                |  32%
   inline R code fragments

  |.........................                                             |  36%
label: unnamed-chunk-2
  |.............................                                         |  41%
  ordinary text without R code

  |................................                                      |  45%
label: unnamed-chunk-3
  |...................................                                   |  50%
  ordinary text without R code

  |......................................                                |  55%
label: unnamed-chunk-4 (with options) 
List of 2
 $ fig.height: num 6
 $ eval      : logi TRUE

  |.........................................                             |  59%
  ordinary text without R code

  |.............................................                         |  64%
label: unnamed-chunk-5 (with options) 
List of 3
 $ fig.height: num 8
 $ fig.width : num 12
 $ eval      : logi TRUE

  |................................................                      |  68%
  ordinary text without R code

  |...................................................                   |  73%
label: unnamed-chunk-6 (with options) 
List of 3
 $ fig.height: num 3
 $ fig.width : num 8
 $ eval      : logi TRUE

  |......................................................                |  77%
  ordinary text without R code

  |.........................................................             |  82%
label: unnamed-chunk-7 (with options) 
List of 3
 $ fig.height: num 2
 $ fig.width : num 8
 $ eval      : logi TRUE

  |............................................................          |  86%
  ordinary text without R code

  |................................................................      |  91%
label: unnamed-chunk-8 (with options) 
List of 2
 $ fig.width: num 8
 $ eval     : logi TRUE

  |...................................................................   |  95%
  ordinary text without R code

  |......................................................................| 100%
label: unnamed-chunk-9 (with options) 
List of 2
 $ fig.width: num 8
 $ eval     : logi TRUE

output file: /tmp/RtmpxBNxzh/Descriptive_Statistics.knit.md

/usr/bin/pandoc +RTS -K512m -RTS /tmp/RtmpxBNxzh/Descriptive_Statistics.knit.md --to latex --from markdown+autolink_bare_uris+tex_math_single_backslash --output /data/Descriptive_Statistics.tex --lua-filter /usr/local/lib/R/site-library/rmarkdown/rmarkdown/lua/pagebreak.lua --lua-filter /usr/local/lib/R/site-library/rmarkdown/rmarkdown/lua/latex-div.lua --self-contained --table-of-contents --toc-depth 4 --highlight-style tango --pdf-engine pdflatex --variable graphics --variable 'geometry:margin=1in' 
Error: LaTeX failed to compile /data/Descriptive_Statistics.tex. See https://yihui.org/tinytex/r/#debugging for debugging tips.
In addition: Warning messages:
1: Removed 899 rows containing non-finite values (stat_bin). 
2: In chisq.test(chi) : Chi-squared approximation may be incorrect
3: In chisq.test(chi) : Chi-squared approximation may be incorrect
4: In (function (survsummary, times, survtable = c("cumevents", "risk.table",  :
  The length of legend.labs should be 8
5: `as.tibble()` was deprecated in tibble 2.0.0.
i Please use `as_tibble()` instead.
i The signature and semantics have changed, see `?as_tibble`.
i The deprecated feature was likely used in the survminer package.
  Please report the issue at <https://github.com/kassambara/survminer/issues>. 
6: `select_()` was deprecated in dplyr 0.7.0.
i Please use `select()` instead.
i The deprecated feature was likely used in the dplyr package.
  Please report the issue at <https://github.com/tidyverse/dplyr/issues>. 
7: In (function (survsummary, times, survtable = c("cumevents", "risk.table",  :
  The length of legend.labs should be 8
Execution halted