Open jpbreed90 opened 1 year ago
Hi,
Could you please send me the output of:
bcftools view -G -H -r 7:36000001-41000000 chr7_berk_60K.vcf.gz
Thanks,
Olivier Delaneau
Le ven. 7 avr. 2023 à 15:07, JP_Breed @.***> a écrit :
Hi there, I am studying shapeit5 and impute5 and I have a question. I successfully performed phasing using shapeit4 on macOS and completed phasing using chunks for SNP arrays.
As I mentioned last time, I have been making efforts to use the shapeit5 version and have finally installed shapeit5 today by creating an environment with ubuntu:20.04 using Docker on macOS (after several twists and turns). Then, I used PLINK to convert a bed file to a vcf file for each chromosome, and converted it to gz format using bgzip and indexed it using bcftools. After that, I proceeded with phasing using options similar to those used in shapeit4, but I keep getting the following error: "ERROR: No variants to be phased!" Below is the phasing log file. How can I solve this problem? [image: image] https://user-images.githubusercontent.com/75996571/230613896-578e2c7e-e860-4fda-a19d-cdfbbefe1e1a.png
P.S. I am experiencing the same error even without using chunks.
— Reply to this email directly, view it on GitHub https://github.com/odelaneau/shapeit5/issues/22, or unsubscribe https://github.com/notifications/unsubscribe-auth/AD4XTIKL3CR6CDXLNJMSE53XAAGSPANCNFSM6AAAAAAWWRWJJ4 . You are receiving this because you are subscribed to this thread.Message ID: @.***>
@jpbreed90 Quick question, do you not have to provide a genetic map to phase (I'm not entirely sure about this)? Assuming you do require a genetic map as one of the inputs, I had a similar issue as well. For me, the "chromosome" field of my genetic map didn't exactly match with the "chromosome" field of my VCF file which is why I got similar error. You might want to look into that and see if it solves your issue.
I'm sorry I'm late. I am currently performing phasing using Shapeit4 on macOS (as Shapeit5 is not installed on macOS), and then using Impute5 on docker with ubuntu:20.04 for imputation. It was difficult to switch between these two environments, so I installed Shapeit4 on docker ubuntu:20.04 and used it, but it also failed to read the variants (ERROR: No variants to be phased!). I'm not sure what the problem is, but I think it might be related to the chromosome number (--region option). I suspect this because when there was an error with the chromosome number in Shapeit4, I encountered the same error.
There was a mention about the genetic_map, so I am sending the genetic_map that I used. I used the example genetic_map distributed by eagle. Also, I have attached the result (aa.txt) for bcftools view -G -H -r 7:36000001-41000000 chr7_berk_60K.vcf.gz.
@jpbreed90 Hi ,I downloaded the docker image from https://github.com/odelaneau/shapeit5/releases/download/v1.0.0/shapeit5_v1.0.0.tar.gz , cat shapeit5_v5.1.0.docker.tar. gz | docker import - shapeit5 docker run -it shapeit5 /bin/bash but it won't start.Do you know how it was solved? thank you~
@jlu0817 Hi Jlu. Have you followed all the installation steps on this site? "https://odelaneau.github.io/shapeit5/docs/installation" I ran it after performing all the steps of installation.
@jpbreed90 Hi, thank you for your reply . I followed that process "https://odelaneau.github.io/shapeit5/docs/installation" , But it still couldn't start . Can you provide commands regarding the docker image start ? Thank you for your time and assistance. I look forward to hearing back from you soon best wishs,
@jlu0817
HI, I am not familiar with the analysis in the Mac environment yet, so it may not be accurate because I am studying.
I first check if the ubuntu environment is working on the docker.
Then use the following code to connect to the ubuntu environment and run shapeit5 (phase_common).
Currently, I'm doing phasing using shapeit4 on a Mac and doing imputation using impute5 in the docker's ubuntu environment. This is because I can't make the program work normally in one environment. I hope my answer will be helpful.
HELLO, did you test region options like chr7:start-end instead 7:start-end? In shapeit4, sometimes matching the chr name can be the solution when no variance error.
Hi,
Can you check if you still have the problem with the v5.1.1 pre-release?
Thanks,
@jlu0817 I have the same problem with you. How did you solve it? Looking forward to your reply!
@BIOJHK I have tested both cases in Shapeit5, but it did not function properly. @yangwukaidi I have not yet tested version 5.1.1, but we have been using Shapeit4 instead of Shapeit5.
Can you please share with me a small example dataset reproducing the error?
@yangwukaidi I ran it using the following command. Hope it can helps you. docker run -v /home/XXX/05.shapeit/docker:/in -w /in shapeit5_v1.0.0:latest ../../usr/bin/SHAPEIT5_phase_common_static_v1.0.0
Just a comment. There are static binaries available in the release if not familiar with docker files. These are straightforward to use.
Hi there, I am studying shapeit5 and impute5 and I have a question. I successfully performed phasing using shapeit4 on macOS and completed phasing using chunks for SNP arrays.
As I mentioned last time, I have been making efforts to use the shapeit5 version and have finally installed shapeit5 today by creating an environment with ubuntu:20.04 using Docker on macOS (after several twists and turns). Then, I used PLINK to convert a bed file to a vcf file for each chromosome, and converted it to gz format using bgzip and indexed it using bcftools. After that, I proceeded with phasing using options similar to those used in shapeit4, but I keep getting the following error: "ERROR: No variants to be phased!" Below is the phasing log file. How can I solve this problem?![image](https://user-images.githubusercontent.com/75996571/230613896-578e2c7e-e860-4fda-a19d-cdfbbefe1e1a.png)
P.S. I am experiencing the same error even without using chunks.