search

aryeelab / hichipper

A preprocessing and QC pipeline for HiChIP data
MIT License
33 stars 12 forks source link

hichipper in CALL mode #78

Open md1587 opened 5 years ago

md1587 commented 5 years ago

Hi, I'm testing hichipper in different modes, and I am having problems with the CALL mode. I am running the latest version (0.7.7) and tested different possibilities: 1) CALL mode with a single valid interaction file as follows hichipper -o test_call --input-vi hicpro/hic_results/data/SRR3467175/SRR3467175.allValidPairs -p GM12878_SMC3.narrowPeak -rf MboI_resfrag_hg19.bed call This generates all the files (non empty) but failed the generation of the QC report. This is the message printed on screen: Mon Oct 07 16:04:19 CEST 2019: Starting hichipper pipeline v0.7.7 Mon Oct 07 16:04:19 CEST 2019: Parsing user parameters Mon Oct 07 16:04:19 CEST 2019: Direct loop call option detected. Anchors removed due to excessive size (likely at ends of chromosomes): 18 Mon Oct 7 16:05:04 CEST 2019: Processing one Mon Oct 7 16:05:04 CEST 2019: Intersecting PETs with anchors Mon Oct 7 16:05:04 CEST 2019: Finished the anchor merging. Mon Oct 7 16:05:04 CEST 2019: Finished the anchor processing. Mon Oct 7 16:31:09 CEST 2019: Intrachromosomal_valid_small=6549671 Mon Oct 7 16:33:46 CEST 2019: Intrachromosomal_valid_med=29890131 Mon Oct 7 16:35:34 CEST 2019: Intrachromosomal_valid_large=11695151 Mon Oct 7 16:35:34 CEST 2019: Total number of anchors used: 53505 Mon Oct 7 16:35:34 CEST 2019: Total number of reads in anchors: 10462287 Mon Oct 7 16:48:17 CEST 2019: Mapped_unique_intra_quality_anchor=797794 Mon Oct 7 16:48:17 CEST 2019: Mapped_unique_intra_quality_anchor_small=329852 Mon Oct 7 16:48:17 CEST 2019: Mapped_unique_intra_quality_anchor_med=405133 Mon Oct 7 16:48:17 CEST 2019: Mapped_unique_intra_quality_anchor_large=62818 Mon Oct 7 16:48:19 CEST 2019: Loop_PETs=405133 hicpro/hic_results/data/SRR3467175/SRR3467175.allValidPairs Mon Oct 07 16:48:19 CEST 2019: ['/usr/local/bin/Rscript', u'test_call/qcReport.R', '/usr/local/lib/python2.7/dist-packages/hichipper', 'test_call', '/home/martina/workspace/test_hichipper', '0.7.7', 'one'] /home/martina/workspace/test_hichipperQuitting from lines 39-101 (qcReport_make.Rmd) Error in .fun(.value[0], ...) : invalid 'type' (character) of argument Calls: ... suppressMessages -> withCallingHandlers -> acast -> cast -> vaggregate Execution halted Processing: one

2) CALL mode with 2 valid interaction files: hichipper -o test_call_3 --input-vi hicpro/hic_results/data/SRR3467175/SRR3467175.allValidPairs,hicpro/hic_results/data/SRR3467176/SRR3467176.allValidPairs -p GM12878_SMC3.narrowPeak -rf MboI_resfrag_hg19.bed call But this is not starting at all. Here is the error message: Tue Oct 08 09:38:00 CEST 2019: Starting hichipper pipeline v0.7.7 Tue Oct 08 09:38:00 CEST 2019: Parsing user parameters Tue Oct 08 09:38:00 CEST 2019: Direct loop call option detected. hicpro/hic_results/data/SRR3467175/SRR3467175.allValidPairs,hicpro/hic_results/data/SRR3467176/SRR3467176.allValidPairs ERROR: in call mode, specify --input-vi

From the --help of hichipper I understood that it was possible to pass a comma-separated list of valid interactions, but it seems that it is not recognising it as multiple inputs. I tried to repeat twice the --input-vi option with each file separately, but it recognised just the last one and then returned the same error as of point 1). Besides, is there a way in CALL mode to keep the sample name instead of "one"? The QC report, along with all the expected output files, is generated without any problem when supplying the manifest file as in the command below: hichipper --out test1 config.yaml where the config.yaml is: peaks:

Thank you very much for any help Cheers Martina

wbszhu commented 4 years ago

Hi， I also have some problems CALL mode with a single valid interaction file ######## Command ####### hichipper --out ./hichipper_results --input-vi /data2/zhanglu/downsample.allValidPairs -p /data2/CTCF_peak.bed -rf /data2/22.genome/AluI.bed call ########## Error ######## Thu Sep 24 10:52:13 CST 2020: Starting hichipper pipeline v0.7.7 Thu Sep 24 10:52:13 CST 2020: Parsing user parameters Thu Sep 24 10:52:13 CST 2020: Direct loop call option detected. Thu Sep 24 10:52:48 CST 2020: Processing one Thu Sep 24 10:52:48 CST 2020: Intersecting PETs with anchors Thu Sep 24 10:52:48 CST 2020: Finished the anchor merging. Thu Sep 24 10:52:48 CST 2020: Something went wrong in determining peaks for anchor inference; rerun with the flag to debug. /data2/zhanglu/downsample.allValidPairs ######### Warning message: In fread(peaksfile, header = FALSE) : File './hichipper_results/userSuppliedPeaks.bed.tmp_pad.bed.tmp' has size 0. Returning a NULL data.table. Error in .find_seqnames_col(df_colnames0, seqnames.field0, prefix) : cannnot find seqnames column Calls: makeGRangesFromDataFrame -> .find_GRanges_cols -> .find_seqnames_col Execution halted Error: The requested file (./hichipper_results/userSuppliedPeaks.bed.tmp_pad.bed.tmprf.tmp) could not be opened. Error message: (No such file or directory). Exiting! /data2/zhanglu/miniconda2/envs/hichipper/lib/python2.7/site-packages/hichipper/interactionsCall_inputFrags.sh: line 35: --keep-temp-files: command not found ERROR: something failed at the individual sample level; check the .log file for more info