Issue: Files Extracting into Different Directory Structure

[poddarharsh15]

Hi @jgust1,

I hope you're doing well!

After downloading the tar file from your page using the following command:

wget https://s3.amazonaws.com/1000g-ont/needLR/needLR_local.tar.gz

tar -xvzf needLR_local.tar.gz

I noticed that the files are being extracted into a path like n/users/jgust1/PROJECTS/needLR/needLR_local/ instead of the expected needLR_local directory with all the required files in it.

Is this extraction path expected, or should it result in a needLR_local directory directly? Could you please take a look and confirm?

Many thanks in advance!

Best regards,
Harsh :)

[jgust1]

Hi Harsh, Thank you for pointing that out! I recently made a small update and must have packaged/re-uploaded the files incorrectly. I will fix that today or tomorrow and let you know when it's done! Best, Gus

…

On Wed, Oct 9, 2024 at 2:59 AM Harsh Poddar ***@***.***> wrote: Hi @jgust1 <https://urldefense.com/v3/__https://github.com/jgust1__;!!K-Hz7m0Vt54!k-TcNOPJUI_SxO09iMI4mUtpRFzMAKWxJe8vCZFOdtpFuhIChhmYQ2_6zXt34fUlItV2Lzy26jHgDZJzI_Y52Uk$> , I hope you're doing well! After downloading the tar file from your page using the following command: wget https://s3.amazonaws.com/1000g-ont/needLR/needLR_local.tar.gz <https://urldefense.com/v3/__https://s3.amazonaws.com/1000g-ont/needLR/needLR_local.tar.gz__;!!K-Hz7m0Vt54!k-TcNOPJUI_SxO09iMI4mUtpRFzMAKWxJe8vCZFOdtpFuhIChhmYQ2_6zXt34fUlItV2Lzy26jHgDZJz_vFpE1E$> tar -xvzf needLR_local.tar.gz I noticed that the files are being extracted into a path like n/users/jgust1/PROJECTS/needLR/needLR_local/ instead of the expected needLR_local directory with all the required files in it. Is this extraction path expected, or should it result in a needLR_local directory directly? Could you please take a look and confirm? Many thanks in advance! Best regards, Harsh :) — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1__;!!K-Hz7m0Vt54!k-TcNOPJUI_SxO09iMI4mUtpRFzMAKWxJe8vCZFOdtpFuhIChhmYQ2_6zXt34fUlItV2Lzy26jHgDZJz8mnWb7k$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED7ZTVK73OBRHTJFSGNLZ2T46LAVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGU3TKNBSHAYDQOA__;!!K-Hz7m0Vt54!k-TcNOPJUI_SxO09iMI4mUtpRFzMAKWxJe8vCZFOdtpFuhIChhmYQ2_6zXt34fUlItV2Lzy26jHgDZJz5SwaVmw$> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[poddarharsh15]

Hi @jgust1,
I have reinstalled the packages and it's working well now :)

[jgust1]

Glad to hear it! Please let me know if you have any other questions. Best, Gus

…

On Mon, Oct 21, 2024 at 3:32 AM Harsh Poddar ***@***.***> wrote: Hi @jgust1 <https://urldefense.com/v3/__https://github.com/jgust1__;!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNd7aAYA_0$> , I have reinstalled the packages and it's working well now :) Thanks again for the tool. — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2426275784__;Iw!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNdwvxHGDA$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED72LR3XUM3A6R2BF3GLZ4TJ25AVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRWGI3TKNZYGQ__;!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNdJnwSY7k$> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[poddarharsh15]

Glad to hear it! Please let me know if you have any other questions.

Best,

Gus

On Mon, Oct 21, 2024 at 3:32 AM Harsh Poddar @.***>

wrote:

Hi @jgust1

https://urldefense.com/v3/__https://github.com/jgust1__;!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNd7aAYA_0$

,

I have reinstalled the packages and it's working well now :) Thanks again

for the tool.

—

Reply to this email directly, view it on GitHub

https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2426275784__;Iw!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNdwvxHGDA$,

or unsubscribe

https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED72LR3XUM3A6R2BF3GLZ4TJ25AVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRWGI3TKNZYGQ__;!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNdJnwSY7k$

.

You are receiving this because you were mentioned.Message ID:

@.***>

Hi Gus,
Quick question about the newer version of needLR—will it be able to detect inheritance patterns if provided with trios (father, mother, and proband)? Specifically, will it show whether the proband's SVs are inherited from the parents or if they are de novo mutations?
I hope I explained that clearly!

Thanks!
Harsh

[jgust1]

Hey Harsh - The next release will include a feature to do that, yes! I hope to have that available by the end of the month. I will message you when it is ready. Best, Gus On Mon, Oct 21, 2024 at 9:26 AM Harsh Poddar ***@***.***> wrote:

…

Glad to hear it! Please let me know if you have any other questions. Best, Gus On Mon, Oct 21, 2024 at 3:32 AM Harsh Poddar *@*.***> wrote: Hi @jgust1 <https://urldefense.com/v3/__https://github.com/jgust1__;!!K-Hz7m0Vt54!maDbeAbVdSX-rL9Ib4jk9Q_mySQ2GgoqRx3qKeFt37E3JCfktY25lysFofeqMmyI4lDvMICvTKy4X1Pv_LnJBsE$> https://urldefense.com/v3/__https://github.com/jgust1__;!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNd7aAYA_0$ , I have reinstalled the packages and it's working well now :) Thanks again for the tool. — Reply to this email directly, view it on GitHub https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2426275784__;Iw!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNdwvxHGDA$ , or unsubscribe https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED72LR3XUM3A6R2BF3GLZ4TJ25AVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRWGI3TKNZYGQ__;!!K-Hz7m0Vt54!hupIPCQyno5nJa71k0fwFI7aXTU-KV86CHXKCCdv-J1_01aJBKIYWFDbsSwJJVRu26y1uuKSP0z3yBNdJnwSY7k$ . You are receiving this because you were mentioned.Message ID: *@*.***> Hi Gus, Quick question about the newer version of needLR—will it be able to detect inheritance patterns if provided with trios (father, mother, and proband)? Specifically, will it show whether the proband's SVs are inherited from the parents or if they are de novo mutations? I hope I explained that clearly! Thanks! Harsh — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2427153374__;Iw!!K-Hz7m0Vt54!maDbeAbVdSX-rL9Ib4jk9Q_mySQ2GgoqRx3qKeFt37E3JCfktY25lysFofeqMmyI4lDvMICvTKy4X1Pvb2_g6dE$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED7ZF6XQWEIIFMHVV7RTZ4UTMJAVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRXGE2TGMZXGQ__;!!K-Hz7m0Vt54!maDbeAbVdSX-rL9Ib4jk9Q_mySQ2GgoqRx3qKeFt37E3JCfktY25lysFofeqMmyI4lDvMICvTKy4X1PvAbLpqac$> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[poddarharsh15]

Hi @jgust1
Any updates about newer version of needLR please let me know?

Thanks.

[jgust1]

Hi Harsh, Great timing! I am actually pushing the update later today! (In about 3 hours) Please let me know if you run into any issues, as you might be the first to actually use the new version. Best, Gus

…

On Thu, Nov 7, 2024 at 7:16 AM Harsh Poddar ***@***.***> wrote: Hi @jgust1 <https://urldefense.com/v3/__https://github.com/jgust1__;!!K-Hz7m0Vt54!hnuPP54sPg9ZlTjN3uumfOnq9aYxfBuKxc9hDq8x912t2GxJEuFfkfFTUmsj5uQMmZATJ4ZA1zAQyjMxiApHf_4$> Any updates about newer version of needLR please let me know? Thanks. — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2462348549__;Iw!!K-Hz7m0Vt54!hnuPP54sPg9ZlTjN3uumfOnq9aYxfBuKxc9hDq8x912t2GxJEuFfkfFTUmsj5uQMmZATJ4ZA1zAQyjMxicJgYos$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED76XANQDTR4Y2PDYK4TZ7NY3JAVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINRSGM2DQNJUHE__;!!K-Hz7m0Vt54!hnuPP54sPg9ZlTjN3uumfOnq9aYxfBuKxc9hDq8x912t2GxJEuFfkfFTUmsj5uQMmZATJ4ZA1zAQyjMxUIammJ8$> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[poddarharsh15]

Hi @jgust1,
I've been trying to work with needLR Trio, but unfortunately, the link seems to be broken. When I follow the instructions, it leads to a "404 - page not found" error. I’ve attached the link below for reference. Could you please take a look when you get a chance?
Thanks!
https://github.com/jgust1/needLR/blob/main/needLR_trio

[poddarharsh15]

Hi Gus,

Quick update—I was able to successfully run the new version of needLR (3.3), and from what I understand, needLR seems to be matching each sample individually from my input list.txt [3 samples] with the 1KGP 150 samples. However, I was expecting that the tool would first merge all my input samples and then match the combined data with the 1KGP samples.

Could you provide some insights on this? Also, just a heads-up, I'll be running about 300 samples for this analysis once the trios version is ready.

Thanks so much!

[jgust1]

Hi Harsh, Sorry about the issue with needLR_trio - It should be fixed now. You'll want to redownload the tar file from AWS to include the needLR_trio_3.3.sh script in the directory (or you can copy the script directly from the GitHib). Re: standard needLR, that is correct - it compares one sample at a time to the pre-merged 150 1KGP samples. Are you interested in looking for SVs that are unique *within* your sample set as well? If so, I could try to help you make a custom script to handle that. Thank you for your enthusiasm for the tool! Please let me know if you run into any issues with needLR_trio - you are probably the first person who will use it. :) Best, Gus

…

On Fri, Nov 8, 2024 at 5:00 AM Harsh Poddar ***@***.***> wrote: Hi Gus, Quick update—I was able to successfully run the new version of needLR (3.3), and from what I understand, needLR seems to be matching each sample individually from my input list.txt [3 samples] with the 1KGP 150 samples. However, I was expecting that the tool would first merge all my input samples and then match the combined data with the 1KGP samples. Could you provide some insights on this? Also, just a heads-up, I'll be running about 300 samples for this analysis once the trios version is ready. Thanks so much! — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2464570685__;Iw!!K-Hz7m0Vt54!ni1Vu10QIjf8_jXr82jfq2rZT0hT1rrkdRbUZQcA68lhHgrq49o8vuJsIUoApm8-sEI7yPPZLm-51V9pLWj7QuU$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED754OSQOTIVX2OCQBETZ7SRU7AVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINRUGU3TANRYGU__;!!K-Hz7m0Vt54!ni1Vu10QIjf8_jXr82jfq2rZT0hT1rrkdRbUZQcA68lhHgrq49o8vuJsIUoApm8-sEI7yPPZLm-51V9pc00N6uw$> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[poddarharsh15]

Hi Gus,

Thanks again for all your help and suggestions! I wanted to share some updates and clarify a few points:

1.Script Typo: I was able to successfully run needLR_trio.sh. Just a heads-up, I noticed a typo in the code at this line:
I've included the corrected command here for reference.

needLR_3.3.sh -f /path/to/proband.vcf -m /path/to/maternal.vcf -f /path/to/paternal.vcf -g /file/path/to/reference/genome.fa

2.Current Work: I have around 300 trio samples, all from Italian patients. Given that we’re focusing on this Italian cohort, I wonder if using samples from 1KGP is strictly necessary. Could you provide your thoughts on this?

3.Unique SV Detection: I’m particularly interested in identifying SVs that are unique within my cohort. Is there a way to do this effectively with needLR?

4.Variant Prioritization: Lastly, I need to prioritize these variants based on their pathogenicity or potential impact on the probands. Any suggestions on how best to approach this with needLR, or if there’s another tool you’d recommend for prioritization?

I really appreciate your help, and I’m looking forward to hearing your thoughts!

Best,
Harsh

[poddarharsh15]

Hi Gus,

Could you please take a look at the results I've generated with one trio example?

UD_AN001_P.tagged_needLR-3.3_trio_UWONT150_RESULTS.txt

[jgust1]

Hi Harsh, Interesting - the data columns all look correct. but the data itself does not. Can you confirm what SV caller you are using? And is this whole genome sequencing or adaptive sampling? Gus

…

On Tue, Nov 12, 2024 at 7:03 AM Harsh Poddar ***@***.***> wrote: Hi Gus, Could you please take a look at the results I've generated with one trio example? UD_AN001_P.tagged_needLR-3.3_trio_UWONT150_RESULTS.txt <https://urldefense.com/v3/__https://github.com/user-attachments/files/17718393/UD_AN001_P.tagged_needLR-3.3_trio_UWONT150_RESULTS.txt__;!!K-Hz7m0Vt54!j0bQb3u6E0M58lnmK2CMCNxGGB0ngqG-n6aI9DgUh_I56L1G2ysFP-XGUGeYFD9WC6NEomXAK3iNbVcVA47dDlc$> — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2470775046__;Iw!!K-Hz7m0Vt54!j0bQb3u6E0M58lnmK2CMCNxGGB0ngqG-n6aI9DgUh_I56L1G2ysFP-XGUGeYFD9WC6NEomXAK3iNbVcVGL2Vx-w$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED73RSJBJXKSHDIKYXPT2AIKELAVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINZQG43TKMBUGY__;!!K-Hz7m0Vt54!j0bQb3u6E0M58lnmK2CMCNxGGB0ngqG-n6aI9DgUh_I56L1G2ysFP-XGUGeYFD9WC6NEomXAK3iNbVcVhfHMnr8$> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[poddarharsh15]

Hi @jgust1
Thank you for your response, actually I am working with WES data , as that’s all we currently have from our patients. For SV calling, I’m using Dysgu, a relatively new tool that operates with a machine learning-based approach.
dysgu

Best,
Harsh

[jgust1]

Ahhhh, I see. needLR is designed for long read sequencing data (short read seq/WES is significantly underpowered to detect SVs). Because all of the back-end 1KGP data is long-read ONT, I don't think it will be feasible to really compare WES to the long read data in a meaningful way, unfortunately. Additionally, needLR is optimized for sniffles2 SV calling, which is specific to long read data. I'm sorry for this misunderstanding! Best, Gus

…

On Tue, Nov 12, 2024 at 11:18 AM Harsh Poddar ***@***.***> wrote: Hi @jgust1 <https://urldefense.com/v3/__https://github.com/jgust1__;!!K-Hz7m0Vt54!jPti94VpAcED0C9YUXc1u5JMWWekIgCqysPXFPUs8Q-vJW5As2AlHtkjfAS52Rk5fw0GeBoZlMcjoQfBp3Qjy38$> Thank you for your response, actually I am working with WES data , as that’s all we currently have from our patients. For SV calling, I’m using Dysgu, a relatively new tool that operates with a machine learning-based approach. dysgu <https://urldefense.com/v3/__https://github.com/kcleal/dysgu__;!!K-Hz7m0Vt54!jPti94VpAcED0C9YUXc1u5JMWWekIgCqysPXFPUs8Q-vJW5As2AlHtkjfAS52Rk5fw0GeBoZlMcjoQfBQVDhh0c$> Best, Harsh — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/jgust1/needLR/issues/1*issuecomment-2471365328__;Iw!!K-Hz7m0Vt54!jPti94VpAcED0C9YUXc1u5JMWWekIgCqysPXFPUs8Q-vJW5As2AlHtkjfAS52Rk5fw0GeBoZlMcjoQfB6KcdGW4$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/A3LED754XPBKCUEA3RXHUKD2AJH77AVCNFSM6AAAAABPUESRU6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINZRGM3DKMZSHA__;!!K-Hz7m0Vt54!jPti94VpAcED0C9YUXc1u5JMWWekIgCqysPXFPUs8Q-vJW5As2AlHtkjfAS52Rk5fw0GeBoZlMcjoQfBY5365-g$> . You are receiving this because you were mentioned.Message ID: ***@***.***>