Question: How to use RIPSeeker?
0
gravatar for xiaocong3333
4 months ago by
xiaocong33330 wrote:

Hi, is there anyone very familiar with RIPSeeker? I have tested it on my RIPseq data, I found some problems which I can't solve. My code as below:

library("RIPSeeker")
library("biomaRt")
library("curl")
library("org.Dm.eg.db")

outDir <- file.path(getwd(), "TEST")
# Parameters setting
binSize <- NULL     
minBinSize <- 11000 
maxBinSize <- 12000
multicore <- TRUE
strandType <- "-"   

biomart <- "ensembl"
biomaRt_dataset <- "dmelanogaster_gene_ensembl"
goAnno <- "org.Dm.eg.db"                    

seekOut.PRC2 <- ripSeek(bamPath = "/home/.../RIPSEQ/test", cNAME = 
                        "Non-...sorted.bam",
                        reverseComplement = TRUE, paired = TRUE,
                        strandType = strandType,
                        uniqueHit = TRUE, assignMultihits = TRUE,
                        rerunWithDisambiguatedMultihits = TRUE,
                        binSize=binSize, minBinSize = minBinSize,
                        maxBinSize = maxBinSize,
                        biomart = biomart,
                        biomaRt_dataset = biomaRt_dataset,
                        multicore=multicore, outDir=outDir)

I run the test data, the result is good, but for the ripRPKM, it reported errors with package “curl” .

When I run my RIPseq data, there are some errors. My library is a directional strand library from fly. I used Hisat2 (an subsitition for Tophat)for mapping, I tried both Dm3/BDGP6 as ref, I changed the strandType with “+”, “-”, “*”, and reverseComplement as TRUE or FALSE.

There are always some problems. When I set strandType = "-/+", reverseComplement as "TRUE/FALSE", the error is: Error in (function (...) : all elements in '...' must be GRanges objects Calls: ripSeek ... as -> asMethod -> do.call -> do.call -> <anonymous> I set the genomeBiuld as 'dm', 'dm3', "dm6", or just delete it, the final result seems no difference.

But when I delete the paired = TRUE and goAnno( If I keep goAnno, it will report less than 2 gene for GO enrichment), but my libraries are paired ends, it seems work. But when I check the RIPregions_annotated.txt, there are only a few information, these are all the results from RIPregions_annotated.txt:

chr2L   22423991        22435020        11030   -       NA      NA      NA      16018   1452.22121486854        379.155319281055        6.45394687429009e-18    6.45394687429009e-18    10573   958.56754306437 -379.422603380833       6.45394687429009e-18    1       NA      NA      NA      NA      NA      NA      NA      NA
chr2R   10239301        10250310        11010   -       NA      NA      NA      597589  54276.9300635786        804.930196349324        9.57741037608529e-33    9.57741037608529e-33    108069  9815.53133514986
        -683.951264684174       9.57741037608529e-33    2       NA      NA      NA      NA      NA      NA      NA      NA
chr2RHet        684171  695205  11035   -       NA      NA      NA      2324    210.602628001812        700.935428402131        8.94887332351163e-15    1.78977466470233e-14    4903    444.313547802447        -699.260953254821       8.94887332351163e-15    3       NA      NA      NA      NA      NA      NA      NA      NA
chr3LHet        2391341 2402360 11020   -       NA      NA      NA      12510   1135.20871143376        7.0982117815557 0.442349878910107       0.442349878910107       9353    848.729582577133        3.01016459563286        0.442349878910107       4       NA      NA      NA      NA      NA      NA      NA      NA
chr3RHet        2470841 2481920 11080   -       NA      NA      NA      5928    535.018050541516        1.33358189119612        0.0200600377053582      0.0200600377053582      7284    657.400722021661        5.73707063985334        0.0200600377053582      5       NA      NA      NA      NA      NA      NA      NA      NA
chrUextra       28523001        28534000        11000   -       NA      NA      NA      9862    896.545454545455        7.23892616977548        0.304860511433238       0.304860511433238       7010    637.272727272727        16.7024751279322        0.304860511433238       6       NA      NA      NA      NA      NA      NA      NA      NA
chrX    7826941 7837995 11055   -       FBgn0029962     CG1402  NA      11171   1010.49298959747        702.925367593253        2.28797076005909e-06    2.36746089331267e-06    9042    817.910447761194        -702.771393425973       2.28797076005909e-06    7       FBgn0029962     7826197 7842679 +       inside  744     744     NearestLocation
chrX    17555341        17566395        11055   -       FBgn0263706     CG43658 NA      11266   1019.08638625057        701.768520799361        2.36746089331267e-06    2.36746089331267e-06    13053   1180.73270013569        -701.847435810435       2.36746089331267e-06    8       FBgn0263706     17456169        17561180        -       overlapStart    5839    5215    NearestLocation
chrXHet 34636   46180   11545   -       NA      NA      NA      833     72.1524469467302        0.896358105384567       0.898471262092674       0.898471262092674       1176    101.862278042443        2.00304273493262        0.898471262092674       9       NA      NA      NA      NA      NA      NA      NA      NA

Thanks in advance!

ADD COMMENTlink modified 4 months ago by WouterDeCoster42k • written 4 months ago by xiaocong33330

Not an answer but unless you really need to use RIPseeker, I would advise to look for alternatives. As far as I can tell development as been stagnating fro a long time now, with the last commit in github being 2 years ago, and pull requests and issues going unanswered from around that time. I also ran into several issues when dealing with non-human genome data and eventually choose to go the DEseq way, with my own custom functions for annotation.

ADD REPLYlink modified 4 months ago • written 4 months ago by A. Domingues2.2k

Thank you Domingues. I used DEseq2 to analyze my RNAseq data. Is it OK just use it for RIPseq data? Did you make any modifications?

ADD REPLYlink written 4 months ago by xiaocong33330

No modification at all, I simply did a IP / input comparison and it worked well - at least that is what my colleagues which did the experimental validation told me.

ADD REPLYlink written 4 months ago by A. Domingues2.2k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1322 users visited in the last hour