small suggestions

[kvittingseerup]

I've been using run.Ted a few times now and I've made a couple observations which results in TED giving uninformative error message. These should be easy for you to implement small test to prevent other people from making the same (stupid) mistakes as me.

Mistake 1: Having gene(s) with all zeros (I think). I had a bulk dataset which gave this error:

[1] "inferred cell percentage"
[1] "correct batch effect"
[1] "every gene has at least one zero. vst transformation is NOT feasible"
[1] "run final sampling"
Error: $ operator is invalid for atomic vectors
In addition: Warning message:
In mclapply(1:N, FUN = function(i) { :
  all scheduled cores encountered errors in user code

After I filtered the data to remove genes with only zeroes the run went fine.

Mistake 2:

The second is that due to a typo I tried running with a data.frame which had NAs in - this also cause TED to fail with an uninformative error message - and took a while to figure out.

Mistake 3:
I make a spelling mistake causing my logic argument to retun an empty vector to be used with the tum.idx argument which naturally caused TED to fail with an uninformative error message

Guess the morale is don't underestimate how strange things users can do :)

[tinyi]

Dear Kristoffer, Thank you very much for your feedback. In theory, Mistake 1 should not happen, as TED allows all zero genes in the input, which will not affect the result. It would help if you can provide me a subset of your samples that generates the error, so I may troubleshoot and make sure the outputs are correct. I will add assertions/alerts to prevent mistake 2 and 3 in the next updates. Best, Tinyi

…

On Fri, Feb 21, 2020 at 4:40 AM Kristoffer Vitting-Seerup < ***@***.***> wrote: I've been using run.Ted a few times now and I've made a couple observations which results in TED giving uninformative error message. These should be easy for you to implement small test to prevent other people from making the same (stupid) mistakes as me. *Mistake 1*: Having gene(s) with all zeros (I think). I had a bulk dataset which gave this error: [1] "inferred cell percentage" [1] "correct batch effect" [1] "every gene has at least one zero. vst transformation is NOT feasible" [1] "run final sampling" Error: $ operator is invalid for atomic vectors In addition: Warning message: In mclapply(1:N, FUN = function(i) { : all scheduled cores encountered errors in user code After I filtered the data to remove genes with only zeroes the run went fine. *Mistake 2*: The second is that due to a typo I tried running with a data.frame which had NAs in - this also cause TED to fail with an uninformative error message - and took a while to figure out. *Mistake 3*: I make a spelling mistake causing my logic argument to retun an empty vector to be used with the tum.idx argument which naturally caused TED to fail with an uninformative error message Guess the morale is don't underestimate how strange things users can do :) — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#5?email_source=notifications&email_token=AB4NHSYG5EHNDUBMBGVLG33RD6OSJA5CNFSM4KY7TVVKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IPH3Y7A>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB4NHS2Z542L636V72QEZ3TRD6OSJANCNFSM4KY7TVVA> .

[tinyi]

Hi Kristoffer , Mistake 2 and 3 are now precluded in the latest release. Feel free to let me know if there are any additional questions/suggestions. Best, Tinyi

…

On Fri, Feb 21, 2020 at 3:25 PM Tinyi Chu ***@***.***> wrote: Dear Kristoffer, Thank you very much for your feedback. In theory, Mistake 1 should not happen, as TED allows all zero genes in the input, which will not affect the result. It would help if you can provide me a subset of your samples that generates the error, so I may troubleshoot and make sure the outputs are correct. I will add assertions/alerts to prevent mistake 2 and 3 in the next updates. Best, Tinyi On Fri, Feb 21, 2020 at 4:40 AM Kristoffer Vitting-Seerup < ***@***.***> wrote: > I've been using run.Ted a few times now and I've made a couple > observations which results in TED giving uninformative error message. These > should be easy for you to implement small test to prevent other people from > making the same (stupid) mistakes as me. > > *Mistake 1*: Having gene(s) with all zeros (I think). I had a bulk > dataset which gave this error: > > [1] "inferred cell percentage" > [1] "correct batch effect" > [1] "every gene has at least one zero. vst transformation is NOT feasible" > [1] "run final sampling" > Error: $ operator is invalid for atomic vectors > In addition: Warning message: > In mclapply(1:N, FUN = function(i) { : > all scheduled cores encountered errors in user code > > After I filtered the data to remove genes with only zeroes the run went > fine. > > *Mistake 2*: > > The second is that due to a typo I tried running with a data.frame which > had NAs in - this also cause TED to fail with an uninformative error > message - and took a while to figure out. > > *Mistake 3*: > I make a spelling mistake causing my logic argument to retun an empty > vector to be used with the tum.idx argument which naturally caused TED > to fail with an uninformative error message > > Guess the morale is don't underestimate how strange things users can do :) > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <#5?email_source=notifications&email_token=AB4NHSYG5EHNDUBMBGVLG33RD6OSJA5CNFSM4KY7TVVKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IPH3Y7A>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AB4NHS2Z542L636V72QEZ3TRD6OSJANCNFSM4KY7TVVA> > . >

[ThomasJamesMitchell]

Nice method - just testing it out at the moment.

I have spent a bit of time struggling with error messages also. One of which came about because I provided cell.type.labels and cell.subtype.labels as factors rather than characters. This took a little unpicking as it was not at all obvious what the issue was. A fix might save others time!

Thanks!

[tinyi]

Hi Thomas, Thank you for your suggestions. I will force a conversion to character in the next update. Best, Tinyi

…

On Tue, Jun 29, 2021 at 10:30 AM Thomas Mitchell ***@***.***> wrote: Nice method - just testing it out at the moment. I have spent a bit of time struggling with error messages also. One of which came about because I provided cell.type.labels and cell.subtype.labels as factors rather than characters. This took a little unpicking as it was not at all obvious what the issue was. A fix might save others time! Thanks! — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB4NHSZJY3YNUYLJEG3IVHTTVHKGTANCNFSM4KY7TVVA> .

[ThomasJamesMitchell]

Thanks Tinyi. BW, Tom From: Tinyi Chu ***@***.***> Sent: 29 June 2021 17:04 To: Danko-Lab/TED ***@***.***> Cc: Thomas Mitchell ***@***.***>; Comment ***@***.***> Subject: Re: [Danko-Lab/TED] small suggestions (#5) [EXT] Hi Thomas, Thank you for your suggestions. I will force a conversion to character in the next update. Best, Tinyi

On Tue, Jun 29, 2021 at 10:30 AM Thomas Mitchell ***@***.***> wrote: Nice method - just testing it out at the moment. I have spent a bit of time struggling with error messages also. One of which came about because I provided cell.type.labels and cell.subtype.labels as factors rather than characters. This took a little unpicking as it was not at all obvious what the issue was. A fix might save others time! Thanks! — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#5 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB4NHSZJY3YNUYLJEG3IVHTTVHKGTANCNFSM4KY7TVVA> .

— You are receiving this because you commented. Reply to this email directly, view it on GitHub [github.com]<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_Danko-2DLab_TED_issues_5-23issuecomment-2D870726290&d=DwMFaQ&c=D7ByGjS34AllFgecYw0iC6Zq7qlm8uclZFI0SqQnqBo&r=HSjcOciWqfX_55HQNNI44Q&m=UMsZVOLf6em-4CQN7JX5JVUf_5wNB2gi7GoapwY1ShE&s=hklaEXauRuS-lk094yKIlk9rz22O7KyIFfqpGw-9HIk&e=>, or unsubscribe [github.com]<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_AH4JS7XJFXWA24FVL7VIW6TTVHVGDANCNFSM4KY7TVVA&d=DwMFaQ&c=D7ByGjS34AllFgecYw0iC6Zq7qlm8uclZFI0SqQnqBo&r=HSjcOciWqfX_55HQNNI44Q&m=UMsZVOLf6em-4CQN7JX5JVUf_5wNB2gi7GoapwY1ShE&s=RzX-NGqakOPffWeV-fWn6T-DPmxKfQV4eMwwrjs4JoU&e=>.

…

-- The Wellcome Sanger Institute is operated by Genome Research Limited, a charity registered in England with number 1021457 and a company registered in England with number 2742969, whose registered office is 215 Euston Road, London, NW1 2BE.