Inquiring about the solution for the missing clone in the input table

[LawrenceH622]

Dear Authors,

Thank you for developing such an excellent tool for clone-specific copy number alteration estimation. I am currently using this tool to analyze my data (CONIPHER and ASCAT converted into refphase output) derived from WES. However, I have encountered an issue during the analysis of the input table processing. Specifically, there is a mismatch between the ci table and the cp table, as they do not share the same clones, which results in the following error. I tried to manually supplement a cn=2 segment for the missing segment in the samples. While the subsequent process does not produce any errors, it generates an unusual output from your plotting script, which seems to be influenced by the supplemented segment.

Could you please clarify how I should handle such an error? Additionally, I would appreciate it if you could provide a solution or suggest an alternative approach to identify and resolve this issue. Thank you in advance for your guidance and assistance.
Error message:

     _____ __    _____ _____ _____ _____
    |  _  |  |  |  _  |  _  |     |  _  |
    |     |  |__|   __|     |   --|     |
    |__|__|_____|__|  |__|__|_____|__|__|
    /\⌒⌒⌒/\
    (⦿   ⦿)
    ( 'Y' )
     (   )
     (   )
     (   )
     (~ ~~~~~~~~~~)
     ( ~ ~~   ~~  )
     ( ~  ~ ~  ~  )
     (~  ~~~~~   ~)
     │ │      │ │
     │ │      │ │
    

Processing files:   0%|          | 0/21 [00:00<?, ?file/s]2025-12-23 14:25:36 - ALPACA - ERROR - An error occurred: Sample names in input table, cp_table and ci_table do not match

        Sample names in input table: ['OM15_P7']
        Sample names in clone proportions table: ['OM15_P13', 'OM15_P7']
        Sample names in confidence intervals table: ['OM15_P13', 'OM15_P7']

Processing files:   0%|          | 0/21 [00:00<?, ?file/s]
Traceback (most recent call last):
  File "/home/u5430259/.conda/envs/alpaca/bin/alpaca", line 7, in <module>
    sys.exit(main())
             ~~~~^^
  File "/home/u5430259/.conda/envs/alpaca/lib/python3.13/site-packages/alpaca/__main__.py", line 41, in main
    run_alpaca()
    ~~~~~~~~~~^^
  File "/home/u5430259/.conda/envs/alpaca/lib/python3.13/site-packages/alpaca/__main__.py", line 119, in run_alpaca
    raise e
  File "/home/u5430259/.conda/envs/alpaca/lib/python3.13/site-packages/alpaca/__main__.py", line 81, in run_alpaca
    SS = SegmentSolution(input_file_name, config, logger)
  File "/home/u5430259/.conda/envs/alpaca/lib/python3.13/site-packages/alpaca/ALPACA_segment_solution_class.py", line 379, in __init__
    validate_inputs(
    ~~~~~~~~~~~~~~~^
        it=self.input_table, cpt=self.cp_table, cit=self.ci_table, t=self.tree
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/home/u5430259/.conda/envs/alpaca/lib/python3.13/site-packages/alpaca/ALPACA_segment_solution_class.py", line 230, in validate_inputs
    raise ValueError(
        f"Sample names in input table, cp_table and ci_table do not match\n{error_msg}"
    )
ValueError: Sample names in input table, cp_table and ci_table do not match

        Sample names in input table: ['OM15_P7']
        Sample names in clone proportions table: ['OM15_P13', 'OM15_P7']
        Sample names in confidence intervals table: ['OM15_P13', 'OM15_P7']

[wlippa]

Hi, and thanks for trying out the tool! In principle, all 3 inputs should have the same samples. Do you know why your 'input table' has one fewer sample? It is especially puzzling in the light of your 'confidence intervals table', where you do have both OM15_P7 and OM15_P13.

For clarity, I will explain the inputs here again, so that we are on the same page.

1. We start with a multi-sample dataset, where each tumour has at least two different samples (biopsies) taken.
2. We call mutations and cluster them with CONPHER, which gives us a phylogenetic tree of clones, and a 'clone proportions' table (cp_table) in clones x samples shape (so CONPHER-derived clones as index, and sample names as columns).
3. We use the sample samples in the ASCAT/RefPhase copy number analysis, which gives us fractional copy numbers in each sample for each locus.
   Note that the samples in points 2 and 3 must necessarily be the same.

Since you mention 'missing segment', I speculate that your samples have different coverage, or alternatively, that they don't have a 'consensus segmentation'.
If you have a difference in coverage, then either add a diploid missing segments (as you seem to be doing) or remove the segments without coverage in all samples. I think the latter is better.

Since you are using refphase, I would expect that your segments start and end in the same locus for all the samples ('consensus/consisents segmentation'), but maybe not?

For example, if you see that sample A has segment 1_1_100 (chromosome 1, starting at 1 and ending at 100) and your sample B has two segments (1_1_50 and 1_51_100), then you need to break the segment in sample A into two, so that the coordinates between samples match.

I suggest you check your RefPhase output and write a small script to correct any issues with segmentation before running ALPACA. If you attach your ALPACA_input_table to this thread, I am happy to take a quick look.

Thanks again for your effort and feedback!

[wlippa]

Hi the '--plot_output_mode pdf --genome-build hg19' option has been added recently, so if your wrapper does not recognise it, the most likely reasons is that your are using older alpaca version with new wrapper. Could you please pull and install the latest version? What does the 'alpaca -v' return?

…

On Wed, 11 Feb 2026 at 03:30, Yuxuan Gao ***@***.***> wrote: *HeadingSouth* left a comment (McGranahanLab/ALPACA-model#25) <#25 (comment)> Dear Authors, I think I found the reason. It seems that the plotting parameter (alpaca run --plot_output_mode) and function (alpaca plot-tumour) are offline temporarily. I found them in README of https://github.com/McGranahanLab/ALPACA-model, but they are not recorded in alpaca help and alpaca run -h. In addition, your example run reports error now: examples/run_example.sh Error: Unknown or unsupported arguments passed to ALPACA: --plot_output_mode pdf --genome-build hg19 I wish plotting function may come back soon, which is definitely helpful for my research! Thank you! — Reply to this email directly, view it on GitHub <#25 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AG7MQ7YDMZROS7Y7DMYIGWT4LKO6VAVCNFSM6AAAAACP2JSJIGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTQOBRHEZDGNJWGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[HeadingSouth]

Dear Authors, Thank you very much for your guidance! A few days later, I noticed that a new version of ALPACA with plotting function had been released. I have now successfully generated the plots using the latest ALPACA model. Your tutorial is detailed and easy to follow, and I am confident this model will greatly support my research. Additionally, I found detailed SNV information for each CONIPHER-predicted clone in the file ${CONIPHER_OUTPUT}/Trees/treeTable.tsv. I was wondering whether it is possible to access the SNV information for each ALPACA-predicted clone as well. My goal is to review the altered SNVs (their total number, position, altered counts, etc.) in each clone and identify potential driver mutations. Or is treeTable.tsv exactly what I should be using for this purpose? Thank you once again! Mirage. ***@***.*** &nbsp;

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "McGranahanLab/ALPACA-model" ***@***.***&gt;; 发送时间:&nbsp;2026年2月18日(星期三) 晚上7:05 ***@***.***&gt;; ***@***.******@***.***&gt;; 主题:&nbsp;Re: [McGranahanLab/ALPACA-model] Inquiring about the solution for the missing clone in the input table (Issue #25) wlippa left a comment (McGranahanLab/ALPACA-model#25) Hi the '--plot_output_mode pdf --genome-build hg19' option has been added recently, so if your wrapper does not recognise it, the most likely reasons is that your are using older alpaca version with new wrapper. Could you please pull and install the latest version? What does the 'alpaca -v' return? On Wed, 11 Feb 2026 at 03:30, Yuxuan Gao ***@***.***&gt; wrote: &gt; *HeadingSouth* left a comment (McGranahanLab/ALPACA-model#25) &gt; <#25 (comment)&gt; &gt; &gt; Dear Authors, &gt; &gt; I think I found the reason. It seems that the plotting parameter (alpaca &gt; run --plot_output_mode) and function (alpaca plot-tumour) are offline &gt; temporarily. I found them in README of &gt; https://github.com/McGranahanLab/ALPACA-model, but they are not recorded &gt; in alpaca help and alpaca run -h. In addition, your example run reports &gt; error now: &gt; examples/run_example.sh &gt; Error: Unknown or unsupported arguments passed to ALPACA: &gt; --plot_output_mode pdf --genome-build hg19 &gt; &gt; I wish plotting function may come back soon, which is definitely helpful &gt; for my research! Thank you! &gt; &gt; — &gt; Reply to this email directly, view it on GitHub &gt; <#25 (comment)&gt;, &gt; or unsubscribe &gt; <https://github.com/notifications/unsubscribe-auth/AG7MQ7YDMZROS7Y7DMYIGWT4LKO6VAVCNFSM6AAAAACP2JSJIGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTQOBRHEZDGNJWGA&gt; &gt; . &gt; You are receiving this because you were mentioned.Message ID: &gt; ***@***.***&gt; &gt; — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***&gt;

[wlippa]

What ALPACA calls 'clones' in the output corresponds to 'clusters' in the conipher. i.e. 'Clone7' = cluster 7 etc. So you need to check which of your SNVs were assigned to which clusters, and then you can relate copy number events to these SNVs. You should check the conipher output (probably the table you mentioned, but I am not 100% sure) and then check which cluster the SNV mutation belongs to. Then check if this cluster has been included in the tree (note that Conipher sometimes rejects clusters which cannot fit the tree). For example, check the pytree_and_bar.pdf plot, included clusters are on the tree, and the rejected are listed in the right lower corner. Alternatively, see which clones are included in ALPACA input or output - it uses only the clusters/clones included in the tree.