Erro after Kinship matrix caculating

[Dkyuan]

Dear Prof. Zhang:

I met errors when performed GWAS using IIIVmrMLM as shown below:

[1] "Single-environment analysis for Pheno1 begins" [1] "Kinship matrix caculating" Error in t(uu) %*% x : 非整合参数 Error : 找不到对象'outSingle'

System: CentOS 8.0.1905 R version: 4.3.1 IIIVmrMLM: the latest version for Linux

The kinship result file " Pheno1_K.csv " seemed correct. So the erro might rise up in the step of "Single-marker genome-wide scanning" (It worked for some phenotypes) Totally, I had 21 phenotypes end with the errors.

I don't know whether the erro means something wrong within the input data or this is because of no significant SNP identified.

[YuanmingZhang65]

这个问题可能是数据的问题，例如，基因型的某一个位点没有变异，或者数据都是缺失

若输入的主成分文件控制群体结构，看看格式是否正确，个体名称是否匹配？

mike @.***

 

------------------ 原始邮件 ------------------ 发件人: "YuanmingZhang65/IIIVmrMLM" @.>; 发送时间: 2024年9月2日(星期一) 下午3:32 @.>; @.***>; 主题: [YuanmingZhang65/IIIVmrMLM] Erro after Kinship matrix caculating (Issue #10)

Dear Prof. Zhang:

I met errors when performed GWAS using IIIVmrMLM as shown below:

[1] "Single-environment analysis for Pheno1 begins" [1] "Kinship matrix caculating" Error in t(uu) %*% x : 非整合参数 Error : 找不到对象'outSingle'

System: CentOS 8.0.1905 R version: 4.3.1 IIIVmrMLM: the latest version for Linux

The kinship result file " Pheno1_K.csv " seemed correct. So the erro might rise up in the step of "Single-marker genome-wide scanning" (It worked for some phenotypes) Totally, I had 21 phenotypes end with the errors.

I don't know whether the erro means something wrong within the input data or this is because of no significant SNP identified.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>

[Dkyuan]

这个问题可能是数据的问题，例如，基因型的某一个位点没有变异，或者数据都是缺失 若输入的主成分文件控制群体结构，看看格式是否正确，个体名称是否匹配？ mike @.   … ------------------ 原始邮件 ------------------ 发件人: "YuanmingZhang65/IIIVmrMLM" @.>; 发送时间: 2024年9月2日(星期一) 下午3:32 @.>; @.>; 主题: [YuanmingZhang65/IIIVmrMLM] Erro after Kinship matrix caculating (Issue #10) Dear Prof. Zhang: I met errors when performed GWAS using IIIVmrMLM as shown below: [1] "Single-environment analysis for Pheno1 begins" [1] "Kinship matrix caculating" Error in t(uu) %*% x : 非整合参数 Error : 找不到对象'outSingle' System: CentOS 8.0.1905 R version: 4.3.1 IIIVmrMLM: the latest version for Linux The kinship result file " Pheno1_K.csv " seemed correct. So the erro might rise up in the step of "Single-marker genome-wide scanning" (It worked for some phenotypes) Totally, I had 21 phenotypes end with the errors. I don't know whether the erro means something wrong within the input data or this is because of no significant SNP identified. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>

感谢章教授的回复，我核查了一下：

基因型数据获取流程： 是根据表型数据的材料名提取并过滤了使得 各个变异 非ref碱基数大于1，maf >= 0.05, missing rate <0.5, 并从该数据用GCTA软件获得PCA，保证了个体匹配， 之后按照 IIIVmrMLM 手册准备好格式

对第1批数据 9 个表型使用 IIIVmrMLM 进行GWAS，有 7 个表型 完成了GWAS获得曼哈顿图、显著点等， 有 2 个 表型数据出现上述错误。其中 1 个完成GWAS的表型的缺失率比出错表型的缺失率甚至更高；

第 2 批数据有 19 个表型，无表型缺失值，均出现上述错误；

第3 批数据 1 个表型，是表型值为 0 或 1的表型，也出现上述错误 ；

上述3 批数据的表型，同样的基因型数据和PCA数据 使用别的GWAS软件运行时没有错误，均可完成GWAS分析获得结果。

[lvzimi]

我也遇到了同样的问题， Error in t(uu) %*% x : non-conformable arguments Error : object 'outSingle' not found 请问您解决了吗？

[YuanmingZhang65]

Error in t(uu) %*% x : non-conformable arguments Error : object 'outSingle' not found

This problem was found in the matching of phenotypic data, genotypic data, and population structure data. To address this problem, please input the correct names of individuals in the phenotype, genotype, and population structure (three) files. If the individual names are correct, there is no problem.

[lvzimi]

Error in t(uu) %*% x : non-conformable argumentst Error : object 'outSingle' not found

This problem was found in the matching of phenotypic data, genotypic data, and population structure data. To address this problem, please input the correct names of individuals in the phenotype, genotype, and population structure (three) files. If the individual names are correct, there is no problem.

感谢章教授的回复！我的问题得到顺利的解决：我的PCA文件id没有修改，并且PCA文件漏掉了第一行。

[Dkyuan]

我也遇到了同样的问题， Error in t(uu) %*% x : non-conformable arguments Error : object 'outSingle' not found 请问您解决了吗？

It still doesn't work for me. I checked again the individuals' names of pheno. geno. and structure(PCA) files with diff (diff geno.acc pca.acc; diff pca.acc pheno.acc ), it showed that names are correct (with no differences among three files).

Especially in the first run with 9 traits, 7 succeeded and 2 failed.

I will try the windows version.