Giters

kevinVervier / SLINGER

Large-scale Machine Learning for gene expression prediction based on genotype only

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Request for information on which allele was counted in the model

holgerman opened this issue · comments

commented

Hi Kevin,

thank you for sharing SLINGER code!
I tried to run SLINGER on my data, however, I did not found information which allele was counted in the model.
Detailed, file https://github.com/kevinVervier/SLINGER/blob/master/model/AllSNPs-elasticNet.txt includes only columns gene, SNP, and beta, but a column informing on the counted allele code is missing.

Could you provide such information?
Thank you very much indeed,
Holger

commented

Hi Holger,

Thank you for your interest in using SLINGER.
I updated another model file with an additional column containing the allele we used for the count (see AllSNPs-elasticNet_withAllele.txt)

Best,
Kevin

commented

Hi Kevin,

thank you very much for the prompt answer and sorry for giving feedback from my side to your answer not earlier!
Following questions came:

First, in the new column, there are 3291 rows with no minor_allele code, e.g. rs17800281. Is this for a certain reason?

Second, I realized in rare cases that I do not have a corresponding allele in my dataset for a certain SNP *(e.g. rs1043099 T vs G or C). This is not unexpected, as my data likely relates to another dbSNP version. However, could you also provide the second allel - this would give the possibility to better filter out such problematic SNPs.

Thank you again!

Best Greetings, Holger