# Linear regression based on genotype-summary statistics

**URL:** <https://discuss.hail.is/t/linear-regression-based-on-genotype-summary-statistics/1563>\
**Category:** Hail Query & hailctl\
**Created:** [July 31, 2020, 9:55am UTC](https://discuss.hail.is/t/linear-regression-based-on-genotype-summary-statistics/1563 "2020-07-31T09:55:37Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Rost](https://avatars.discourse-cdn.com/v4/letter/r/6de8d8/32.png) [@Rost](https://discuss.hail.is/u/Rost)\
**Post date:** [July 31, 2020, 9:55am UTC](https://discuss.hail.is/t/linear-regression-based-on-genotype-summary-statistics/1563/1 "2020-07-31T09:55:37Z")

</div>

In my study, I don’t have access to particular genotypes. There are only summary statistics per phenotype. Is it possible to do linear regression inside hail based only on the summary statistics?

```auto
Entry fields:
    'non_miss': int64
    'hom_ref': int64
    'hom_var': int64
    'het': int64
----------------------------------------
Column key: ['pheno']

```

---

<div class="post-metadata">

**Author:** ![tpoterba](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/tpoterba/32/61_2.png) [@tpoterba](https://discuss.hail.is/u/tpoterba)\
**Post date:** [July 31, 2020, 4:49pm UTC](https://discuss.hail.is/t/linear-regression-based-on-genotype-summary-statistics/1563/2 "2020-07-31T16:49:46Z")

</div>

it’s certainly possible to do a linear regression on any numeric entry field, but scientifically this might not be what you want. This seems more like a meta-analysis task, yes? @kumarveerapen?

---

<div class="post-metadata">

**Author:** ![kumarveerapen](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.hail.is/kumarveerapen/32/355_2.png) [@kumarveerapen](https://discuss.hail.is/u/kumarveerapen)\
**Post date:** [August 7, 2020, 2:01am UTC](https://discuss.hail.is/t/linear-regression-based-on-genotype-summary-statistics/1563/3 "2020-08-07T02:01:17Z")

</div>

Thanks for tagging me, Tim! 🙂

Hi, Rost! To answer your question, yes, this would be a meta-analysis. This page may be helpful for your needs  
[https://hail.is/docs/0.2/methods/stats.html](https://hail.is/docs/0.2/methods/stats.html)