# Day 5 **URL:** https://isgw-forum.colorado.edu/c/isg-workshops-2026/day-5/34.md [Latest](https://isgw-forum.colorado.edu/latest.md) · [Categories](https://isgw-forum.colorado.edu/categories.md) · [Tags](https://isgw-forum.colorado.edu/tags.md) --- ## [About the Day 5 category](https://isgw-forum.colorado.edu/t/about-the-day-5-category/745)
**Author:** [@JML](https://isgw-forum.colorado.edu/u/JML)\ **Replies:** 0\ **Last updated:** [May 26, 2026, 5:12pm UTC](https://isgw-forum.colorado.edu/t/about-the-day-5-category/745 "2026-05-26T17:12:53Z")
prediction theory, ROC analysis, polygenic risk scores, SBayesRC, G×E applications --- ## [Asking questions about BLUP](https://isgw-forum.colorado.edu/t/asking-questions-about-blup/828)
**Author:** [@xueyingqin](https://isgw-forum.colorado.edu/u/xueyingqin)\ **Replies:** 0\ **Last updated:** [June 9, 2026, 2:34am UTC](https://isgw-forum.colorado.edu/t/asking-questions-about-blup/828 "2026-06-09T02:34:11Z")
Hi Professors, “I have a general question regarding ‘Part 3: Conventional methods for PGS prediction.’ On page 8, it states that BLUP is a linear mixed-effects model. Is this mixed model the same as the model used in GC… --- ## [How to decide best hyperparameters](https://isgw-forum.colorado.edu/t/how-to-decide-best-hyperparameters/827)
**Author:** [@helenl](https://isgw-forum.colorado.edu/u/helenl)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 6:47pm UTC](https://isgw-forum.colorado.edu/t/how-to-decide-best-hyperparameters/827 "2026-06-08T18:47:10Z")
For some Bayesian PRS methods, how do users decide which hyperparameter values to specify or begin with, such as the shrinkage parameter? Some tools might suggest performing a grid search, but from a method development p… --- ## [Cross Validation](https://isgw-forum.colorado.edu/t/cross-validation/826)
**Author:** [@anonymous](https://isgw-forum.colorado.edu/u/anonymous)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 5:15pm UTC](https://isgw-forum.colorado.edu/t/cross-validation/826 "2026-06-08T17:15:54Z")
How many folds are ideal for cross validation? Replying to “How many folds are ideal for cross validation?”: Often people use 5- or 10-fold depending on sample size. 10-fold CV will be more computationally expensive … --- ## [Tuning Sets](https://isgw-forum.colorado.edu/t/tuning-sets/825)
**Author:** [@anonymous](https://isgw-forum.colorado.edu/u/anonymous)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 5:15pm UTC](https://isgw-forum.colorado.edu/t/tuning-sets/825 "2026-06-08T17:15:07Z")
Why use a tuning set at all? Replying to “Why use a tuning set at all?”: to prevent overfitting that you get when you select the best p-threshold Replying to “Why use a tuning set at all?”: Okay, thanks Replyin… --- ## [Using GWAS Results When 23andMe Data Are Restricted](https://isgw-forum.colorado.edu/t/using-gwas-results-when-23andme-data-are-restricted/824)
**Author:** [@anonymous](https://isgw-forum.colorado.edu/u/anonymous)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 5:14pm UTC](https://isgw-forum.colorado.edu/t/using-gwas-results-when-23andme-data-are-restricted/824 "2026-06-08T17:14:10Z")
What is the recommended approach when full GWAS summary statistics are available for only part of the sample, while for another part (e.g., 23andMe participants) only the top 10,000 SNPs are provided? How should this be … --- ## [Sample Size for a PGS Target Sample](https://isgw-forum.colorado.edu/t/sample-size-for-a-pgs-target-sample/823)
**Author:** [@anonymous](https://isgw-forum.colorado.edu/u/anonymous)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 5:11pm UTC](https://isgw-forum.colorado.edu/t/sample-size-for-a-pgs-target-sample/823 "2026-06-08T17:11:07Z")
what is generally considered a good sample size for a PGS target sample population? Replying to “what is generally considered a good sample size fo…”: depends on the heritability of the trait and sample size of the G… --- ## [Combining Polygenic Scores](https://isgw-forum.colorado.edu/t/combining-polygenic-scores/822)
**Author:** [@anonymous](https://isgw-forum.colorado.edu/u/anonymous)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 5:10pm UTC](https://isgw-forum.colorado.edu/t/combining-polygenic-scores/822 "2026-06-08T17:10:17Z")
Are there methods designed for combining the polygenic scores for possible improved prediction? Replying to “Are there methods designed for combining the polyg…”: There are methods for doing this - you can also just … --- ## [Priors and Posterior inference](https://isgw-forum.colorado.edu/t/priors-and-posterior-inference/821)
**Author:** [@anonymous](https://isgw-forum.colorado.edu/u/anonymous)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 5:09pm UTC](https://isgw-forum.colorado.edu/t/priors-and-posterior-inference/821 "2026-06-08T17:09:35Z")
Can you define “Priors” and “Posterior inference” in a bit more detail, in the context of making a PRS for a specificic phenotype example? Replying to “Can you define “Priors” and “Posterior inference” …”: The priors… --- ## [PGS and Family designs](https://isgw-forum.colorado.edu/t/pgs-and-family-designs/820)
**Author:** [@juanjose.madrid1](https://isgw-forum.colorado.edu/u/juanjose.madrid1)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 3:58pm UTC](https://isgw-forum.colorado.edu/t/pgs-and-family-designs/820 "2026-06-08T15:58:29Z")
When working with twin samples and PGS, what is the preferred approach for accounting for relatedness? Is it sufficient to model family structure using a random effect (e.g., 1 | FAMID), or is it preferable to account fo… --- ## [Large effects for Diabetes and Rheumatoid Arthritis at chromosome 6](https://isgw-forum.colorado.edu/t/large-effects-for-diabetes-and-rheumatoid-arthritis-at-chromosome-6/819)
**Author:** [@pietro.bassani](https://isgw-forum.colorado.edu/u/pietro.bassani)\ **Replies:** 0\ **Last updated:** [June 8, 2026, 6:35am UTC](https://isgw-forum.colorado.edu/t/large-effects-for-diabetes-and-rheumatoid-arthritis-at-chromosome-6/819 "2026-06-08T06:35:22Z")
Hello, I was wondering if the large effects reported here have anything to do with the MHC located in at chromosome 6.