# \#scvi

**URL:** https://discourse.scverse.org/tag/scvi/10.md

[Latest](https://discourse.scverse.org/latest.md) · [Categories](https://discourse.scverse.org/categories.md) · [Tags](https://discourse.scverse.org/tags.md)

---

## [Can scVI take as input SCT-normalized data?](https://discourse.scverse.org/t/can-scvi-take-as-input-sct-normalized-data/4042)

<div class="topic-metadata">

**Author:** [@gboscagli](https://discourse.scverse.org/u/gboscagli)\
**Replies:** 2\
**Last updated:** [September 8, 2026, 8:53am UTC](https://discourse.scverse.org/t/can-scvi-take-as-input-sct-normalized-data/4042 "2026-09-08T08:53:42Z")

</div>

Hi all, i am trying to perform scVI integration on a merged dataset processed in R and normalized with scTransform. I want to use native Python implementation rather than Seurat wrapper, since i want to play around with…

---

## [Allowed downstream analysis with get\_normalized\_expression?](https://discourse.scverse.org/t/allowed-downstream-analysis-with-get-normalized-expression/4039)

<div class="topic-metadata">

**Author:** [@gboscagli](https://discourse.scverse.org/u/gboscagli)\
**Replies:** 6\
**Last updated:** [July 30, 2026, 9:06am UTC](https://discourse.scverse.org/t/allowed-downstream-analysis-with-get-normalized-expression/4039 "2026-07-30T09:06:34Z")

</div>

Greetings all, Hope I won’t be redundant with this issue. I am interested in finding an integration tool which returns a full gene-space batch-corrected matrix to use for downstream analysis, and I bumped into scVI’s g…

---

## [SCVI view\_anndata\_setup causes AttributeError with rich.pretty](https://discourse.scverse.org/t/scvi-view-anndata-setup-causes-attributeerror-with-rich-pretty/4022)

<div class="topic-metadata">

**Author:** [@ggruenhagen3](https://discourse.scverse.org/u/ggruenhagen3)\
**Replies:** 1\
**Last updated:** [July 8, 2026, 6:45am UTC](https://discourse.scverse.org/t/scvi-view-anndata-setup-causes-attributeerror-with-rich-pretty/4022 "2026-07-08T06:45:06Z")

</div>

Hello, I was trying to look at my scvi model using view\_anndata\_setup, but got the error below that says ‘rich has no attribute pretty’. I was actually able to do everything else, like train my model completely fine. Eve…

---

## [Getting normalized expression](https://discourse.scverse.org/t/getting-normalized-expression/3976)

<div class="topic-metadata">

**Author:** [@Kray](https://discourse.scverse.org/u/Kray)\
**Replies:** 13\
**Last updated:** [May 13, 2026, 9:54am UTC](https://discourse.scverse.org/t/getting-normalized-expression/3976 "2026-05-13T09:54:49Z")

</div>

Hi, I am working on a big sized scRNAseq atlas with 2 million cells. I want to get the normalized expression. However, since it returns a data frame/numpy array, I run out of memory every time I am trying to retrieve th…

---

## [Integration challenges in a multi-dataset single cell type maturation atlas](https://discourse.scverse.org/t/integration-challenges-in-a-multi-dataset-single-cell-type-maturation-atlas/3969)

<div class="topic-metadata">

**Author:** [@grad\_student\_01](https://discourse.scverse.org/u/grad_student_01)\
**Replies:** 2\
**Last updated:** [April 24, 2026, 3:40pm UTC](https://discourse.scverse.org/t/integration-challenges-in-a-multi-dataset-single-cell-type-maturation-atlas/3969 "2026-04-24T15:40:41Z")

</div>

Hi scverse community, I’m building a single-cell maturation atlas integrating 23 publicly available and in-house datasets spanning embryonic day 7.75 through postnatal day 84 in mice (~58,000 cells after QC). I’m runni…

---

## [ResolVI Corrected Counts for Segmented Visium HD Data](https://discourse.scverse.org/t/resolvi-corrected-counts-for-segmented-visium-hd-data/3948)

<div class="topic-metadata">

**Author:** [@dow527](https://discourse.scverse.org/u/dow527)\
**Replies:** 3\
**Last updated:** [April 23, 2026, 4:40pm UTC](https://discourse.scverse.org/t/resolvi-corrected-counts-for-segmented-visium-hd-data/3948 "2026-04-23T16:40:50Z")

</div>

Hi, I’m trying to understand the correct method for running ResolVI on Visium HD data that has been segmented using proseg as opposed to 8μM bins used in the reproducibility jupyter notebooks. I’m currently following t…

---

## [Correct way to run SOLO for doublet detection](https://discourse.scverse.org/t/correct-way-to-run-solo-for-doublet-detection/3949)

<div class="topic-metadata">

**Author:** [@shevo33](https://discourse.scverse.org/u/shevo33)\
**Replies:** 2\
**Last updated:** [April 23, 2026, 9:29am UTC](https://discourse.scverse.org/t/correct-way-to-run-solo-for-doublet-detection/3949 "2026-04-23T09:29:07Z")

</div>

Hi everyone, I am trying to run SOLO for doublet detection. However, I am not sure what is the correct way. At first, I thought I have to run Solo after sc.pp.highly\_variable\_genes(adata), but later I realized it makes…

---

## [How to use GPU partitions (MIGs)?](https://discourse.scverse.org/t/how-to-use-gpu-partitions-migs/3951)

<div class="topic-metadata">

**Author:** [@shevo33](https://discourse.scverse.org/u/shevo33)\
**Replies:** 1\
**Last updated:** [March 28, 2026, 1:13pm UTC](https://discourse.scverse.org/t/how-to-use-gpu-partitions-migs/3951 "2026-03-28T13:13:33Z")

</div>

Hi everyone, I am trying to run scvi as followed, but I wan to speed up by using several GPUs. However, my cluster only gives me MIG partitions from GPUs. Even tho I can request several, when feeding the MIGs to scvi…

---

## [Unstable result](https://discourse.scverse.org/t/unstable-result/1090)

<div class="topic-metadata">

**Author:** [@GwangWooKim](https://discourse.scverse.org/u/GwangWooKim)\
**Replies:** 2\
**Last updated:** [March 27, 2026, 3:54pm UTC](https://discourse.scverse.org/t/unstable-result/1090 "2026-03-27T15:54:08Z")

</div>

Hello, I’m going to analyze the single-cell RNA sequencing data using the scVI framework. After training the scVI model, I found that the result of model.differential\_expression is unstable (specifically, it varies from…

---

## [Regarding resolvi reproducibility scripts and fov for xenium](https://discourse.scverse.org/t/regarding-resolvi-reproducibility-scripts-and-fov-for-xenium/3939)

<div class="topic-metadata">

**Author:** [@jairaj-mathur](https://discourse.scverse.org/u/jairaj-mathur)\
**Replies:** 2\
**Last updated:** [March 2, 2026, 8:11pm UTC](https://discourse.scverse.org/t/regarding-resolvi-reproducibility-scripts-and-fov-for-xenium/3939 "2026-03-02T20:11:56Z")

</div>

Hi, I was going through the resolvi reproducibilty scripts, and for xenium, it seems like fov as batch was used. I was wondering how you were able to get that information in your anndata object, as the table from spatial…

---

## [Is older advice for estimating optimal number of epochs for model training of scVI still recommended?](https://discourse.scverse.org/t/is-older-advice-for-estimating-optimal-number-of-epochs-for-model-training-of-scvi-still-recommended/3932)

<div class="topic-metadata">

**Author:** [@timslittle](https://discourse.scverse.org/u/timslittle)\
**Replies:** 6\
**Last updated:** [March 1, 2026, 8:13am UTC](https://discourse.scverse.org/t/is-older-advice-for-estimating-optimal-number-of-epochs-for-model-training-of-scvi-still-recommended/3932 "2026-03-01T08:13:51Z")

</div>

Hi Team, Earlier docs for scVI mention that the number of epochs should scale downwards with number of obs/cells. E.g: “(\[For\] \> 100,000 cells) you may only need ~10 epochs.” I cannot find similar advice on the modern d…

---

## [Scvi-tools not allowing me to download cellxgene census data](https://discourse.scverse.org/t/scvi-tools-not-allowing-me-to-download-cellxgene-census-data/3926)

<div class="topic-metadata">

**Author:** [@igrisch](https://discourse.scverse.org/u/igrisch)\
**Replies:** 2\
**Last updated:** [January 28, 2026, 8:21pm UTC](https://discourse.scverse.org/t/scvi-tools-not-allowing-me-to-download-cellxgene-census-data/3926 "2026-01-28T20:21:22Z")

</div>

Hi there, I am trying to follow the tutorial here: I basically want to take the census data and query my own dataset against all the census data. I am running into the following error when I try to do the tutorial’s s…

---

## [Imputation/Denoise output](https://discourse.scverse.org/t/imputation-denoise-output/3913)

<div class="topic-metadata">

**Author:** [@Huiyueli](https://discourse.scverse.org/u/Huiyueli)\
**Replies:** 1\
**Last updated:** [January 15, 2026, 7:22am UTC](https://discourse.scverse.org/t/imputation-denoise-output/3913 "2026-01-15T07:22:32Z")

</div>

When using the code below for data denoise/imputation, how to set the library\_size in get\_normalized\_expression()? to obtain the output in count-scale, rather than normalized output? scvi.model.SCVI.setup\_anndata(adata,…

---

## [Predict cell type with scANVI for spatial transcriptomics data (Xenium)](https://discourse.scverse.org/t/predict-cell-type-with-scanvi-for-spatial-transcriptomics-data-xenium/3867)

<div class="topic-metadata">

**Author:** [@spatts14](https://discourse.scverse.org/u/spatts14)\
**Replies:** 7\
**Last updated:** [December 28, 2025, 2:31pm UTC](https://discourse.scverse.org/t/predict-cell-type-with-scanvi-for-spatial-transcriptomics-data-xenium/3867 "2025-12-28T14:31:08Z")

</div>

Hi, Thanks so much for the amazing tools! I was hoping someone could help me. I have an in-house dataset of Xenium 5k data and I would like to predict the cell types using a reference dataset. As such, I am also tryin…

---

## [Issue with scVI VAE reconstruction: High sensitivity to library size and poor rank preservation](https://discourse.scverse.org/t/issue-with-scvi-vae-reconstruction-high-sensitivity-to-library-size-and-poor-rank-preservation/3883)

<div class="topic-metadata">

**Author:** [@hamidsta](https://discourse.scverse.org/u/hamidsta)\
**Replies:** 1\
**Last updated:** [December 9, 2025, 2:58pm UTC](https://discourse.scverse.org/t/issue-with-scvi-vae-reconstruction-high-sensitivity-to-library-size-and-poor-rank-preservation/3883 "2025-12-09T14:58:30Z")

</div>

Hello :smiley: , I am running scVI VAE tool, with the goal of analyzing both the learned cell embeddings and the reconstructed gene expression. ( i have merfish sample ) I want to verify if the gene ranking is preser…

---

## [Scvi 1.4.0. model.differential\_expression outputs](https://discourse.scverse.org/t/scvi-1-4-0-model-differential-expression-outputs/3893)

<div class="topic-metadata">

**Author:** [@cagla](https://discourse.scverse.org/u/cagla)\
**Replies:** 1\
**Last updated:** [December 7, 2025, 7:21am UTC](https://discourse.scverse.org/t/scvi-1-4-0-model-differential-expression-outputs/3893 "2025-12-07T07:21:00Z")

</div>

Hello, I am using the scvi version 1.4.0.post1. The model.differential\_expression function gives proba\_m1, proba\_m2, bayes\_factor, scale1, scale2, raw\_mean1, raw\_mean2, non\_zeros\_proportion1, non\_zeros\_proportion2, raw\_n…

---

## [Corrected count for resolvi](https://discourse.scverse.org/t/corrected-count-for-resolvi/3870)

<div class="topic-metadata">

**Author:** [@ynanli](https://discourse.scverse.org/u/ynanli)\
**Replies:** 1\
**Last updated:** [November 25, 2025, 7:42pm UTC](https://discourse.scverse.org/t/corrected-count-for-resolvi/3870 "2025-11-25T19:42:09Z")

</div>

Hi, I am wondering what would be the right way of generating a corrected count matrix for downstream analysis from resolvi. The tutorial uses the median of px\_rate as generated\_expression, while the preprint computed c…

---

## [Question about complex experiment](https://discourse.scverse.org/t/question-about-complex-experiment/3864)

<div class="topic-metadata">

**Author:** [@dackjames](https://discourse.scverse.org/u/dackjames)\
**Replies:** 1\
**Last updated:** [November 25, 2025, 7:39pm UTC](https://discourse.scverse.org/t/question-about-complex-experiment/3864 "2025-11-25T19:39:01Z")

</div>

Hi everyone, I’m starting a fairly complex experiment that includes 10x single-cell, single-nucleus, and 10x multiome (snRNA-seq + scATAC-seq from the same nucleus), with control and treatment groups spread across all t…

---

## [Clarification on the random variable W in the decoder](https://discourse.scverse.org/t/clarification-on-the-random-variable-w-in-the-decoder/3860)

<div class="topic-metadata">

**Author:** [@jcara514](https://discourse.scverse.org/u/jcara514)\
**Replies:** 2\
**Last updated:** [November 11, 2025, 1:33pm UTC](https://discourse.scverse.org/t/clarification-on-the-random-variable-w-in-the-decoder/3860 "2025-11-11T13:33:45Z")

</div>

Hello, So I’ve been doing a deep dive into the technical aspects of this software for my dissertation project. Something that I noticed while going through the decoder algorithm was the following: w is defined as Gamma(…

---

## [Training not using the full GPU](https://discourse.scverse.org/t/training-not-using-the-full-gpu/3857)

<div class="topic-metadata">

**Author:** [@mys\_721tx](https://discourse.scverse.org/u/mys_721tx)\
**Replies:** 3\
**Last updated:** [November 10, 2025, 2:35pm UTC](https://discourse.scverse.org/t/training-not-using-the-full-gpu/3857 "2025-11-10T14:35:50Z")

</div>

So I have been running training couple models recently and I encountered this one problem. scvi has no problem detecting CUDA or the GPU. But the problem is that it is using about 25% of GPU and 20% of VRAMs. Have any of…

---

## [Question about get\_normalized\_expression computation](https://discourse.scverse.org/t/question-about-get-normalized-expression-computation/3842)

<div class="topic-metadata">

**Author:** [@racng](https://discourse.scverse.org/u/racng)\
**Replies:** 3\
**Last updated:** [November 5, 2025, 8:14am UTC](https://discourse.scverse.org/t/question-about-get-normalized-expression-computation/3842 "2025-11-05T08:14:27Z")

</div>

I have trained a TOTALVI model and I am using the get\_normalized\_expression function to get imputed values for one gene and one protein. It is using mostly CPUs (128 threads) and not the GPU (Volatile GPU-Util = 8%, 2391…

---

## [Same scVI model with different values on mac silicon and linux?!](https://discourse.scverse.org/t/same-scvi-model-with-different-values-on-mac-silicon-and-linux/3846)

<div class="topic-metadata">

**Author:** [@munta](https://discourse.scverse.org/u/munta)\
**Replies:** 1\
**Last updated:** [November 4, 2025, 7:09am UTC](https://discourse.scverse.org/t/same-scvi-model-with-different-values-on-mac-silicon-and-linux/3846 "2025-11-04T07:09:02Z")

</div>

Hi. I have used scvi model to integrate data from different batches. I used HPC cluster and gpu for that. when I checked the data on cluster adata.layers\['scvi\_norm'\] = vae.get\_normalized\_expression(adata=adata,librar…

---

## [Excluding Ig and ribosomal genes from HVG selection in scVI, best practice?](https://discourse.scverse.org/t/excluding-ig-and-ribosomal-genes-from-hvg-selection-in-scvi-best-practice/3833)

<div class="topic-metadata">

**Author:** [@marencc](https://discourse.scverse.org/u/marencc)\
**Replies:** 1\
**Last updated:** [October 22, 2025, 9:15am UTC](https://discourse.scverse.org/t/excluding-ig-and-ribosomal-genes-from-hvg-selection-in-scvi-best-practice/3833 "2025-10-22T09:15:23Z")

</div>

Hi everyone, I’m working with single-cell RNA-seq data from CD45⁺ immune cells (mostly lymphoid lineages) and integrating multiple batches using scVI, which so far has given the best batch correction results. We’re now…

---

## [How to handle batch effects within the query dataset when using scArches + SCVI?](https://discourse.scverse.org/t/how-to-handle-batch-effects-within-the-query-dataset-when-using-scarches-scvi/3728)

<div class="topic-metadata">

**Author:** [@ashenflower](https://discourse.scverse.org/u/ashenflower)\
**Replies:** 1\
**Last updated:** [July 31, 2025, 12:14pm UTC](https://discourse.scverse.org/t/how-to-handle-batch-effects-within-the-query-dataset-when-using-scarches-scvi/3728 "2025-07-31T12:14:23Z")

</div>

Hello everyone, I’m new to scArches and currently exploring how it works in combination with SCVI, particularly for integrating new datasets into a reference atlas. I’m following this tutorial, where a SCVI model is fi…

---

## [Implementing Warmup to a component based on epoch](https://discourse.scverse.org/t/implementing-warmup-to-a-component-based-on-epoch/3736)

<div class="topic-metadata">

**Author:** [@dbdimitrov](https://discourse.scverse.org/u/dbdimitrov)\
**Replies:** 2\
**Last updated:** [July 31, 2025, 6:43am UTC](https://discourse.scverse.org/t/implementing-warmup-to-a-component-based-on-epoch/3736 "2025-07-31T06:43:05Z")

</div>

Hi, This might be a silly question, but I’m trying to implement a warmup schedule for one of the components of my model. I’ve implemented some make-shift solution, but was hoping to avoid digging through the codebase i…

---

## [Deconvolute spatial transcriptomics (visium) with unannotated single-cell (10x)](https://discourse.scverse.org/t/deconvolute-spatial-transcriptomics-visium-with-unannotated-single-cell-10x/3694)

<div class="topic-metadata">

**Author:** [@cell\_kc](https://discourse.scverse.org/u/cell_kc)\
**Replies:** 3\
**Last updated:** [July 10, 2025, 8:10am UTC](https://discourse.scverse.org/t/deconvolute-spatial-transcriptomics-visium-with-unannotated-single-cell-10x/3694 "2025-07-10T08:10:36Z")

</div>

I have a spatial transcriptomics dataset (10x visium) of a tumour microarray and some of the samples have matched single-cell RNA-seq data (10x). I would like to annotate my spatial data with the single-cell data. Howeve…

---

## [Does MultiVI support using tile matrix directly instead of peak calling for ATAC input?](https://discourse.scverse.org/t/does-multivi-support-using-tile-matrix-directly-instead-of-peak-calling-for-atac-input/3701)

<div class="topic-metadata">

**Author:** [@Captain\_Pam](https://discourse.scverse.org/u/Captain_Pam)\
**Replies:** 2\
**Last updated:** [July 9, 2025, 4:26pm UTC](https://discourse.scverse.org/t/does-multivi-support-using-tile-matrix-directly-instead-of-peak-calling-for-atac-input/3701 "2025-07-09T16:26:22Z")

</div>

Hi all, I’m working with 10x multiome data and have processed the ATAC and RNA modalities separately: For ATAC, I used tile matrix (e.g., 5kb bins) and trained a peakVI model. For RNA, I used scVI. Now, I’m planning …

---

## [DestVI.from\_rna\_model RuntimeError: Error(s) in loading state\_dict for FCLayers](https://discourse.scverse.org/t/destvi-from-rna-model-runtimeerror-error-s-in-loading-state-dict-for-fclayers/3711)

<div class="topic-metadata">

**Author:** [@aleydace](https://discourse.scverse.org/u/aleydace)\
**Replies:** 1\
**Last updated:** [July 9, 2025, 4:20pm UTC](https://discourse.scverse.org/t/destvi-from-rna-model-runtimeerror-error-s-in-loading-state-dict-for-fclayers/3711 "2025-07-09T16:20:58Z")

</div>

Hi. I am getting an error running DestVi (scvi-tools 1.3.0). I am basically importing a single cell reference AnnData, subsetting to keep just variable genes, and then finding shared genes with my spatial AnnData. I then…

---

## [Shared cell types not mixing when integrating datasets from different species](https://discourse.scverse.org/t/shared-cell-types-not-mixing-when-integrating-datasets-from-different-species/3685)

<div class="topic-metadata">

**Author:** [@sandcell10](https://discourse.scverse.org/u/sandcell10)\
**Replies:** 4\
**Last updated:** [June 19, 2025, 7:02pm UTC](https://discourse.scverse.org/t/shared-cell-types-not-mixing-when-integrating-datasets-from-different-species/3685 "2025-06-19T19:02:10Z")

</div>

Hello, I am attempting to integrate datasets of the same tissue from different species. Some of the datasets are single-cell while some are single-nucleus. I thought that SCVI would work well for this task, however, as …

---

## [Domain adaptation to pre-train batch correction model using paired data](https://discourse.scverse.org/t/domain-adaptation-to-pre-train-batch-correction-model-using-paired-data/3633)

<div class="topic-metadata">

**Author:** [@Valerie](https://discourse.scverse.org/u/Valerie)\
**Replies:** 12\
**Last updated:** [May 27, 2025, 6:44am UTC](https://discourse.scverse.org/t/domain-adaptation-to-pre-train-batch-correction-model-using-paired-data/3633 "2025-05-27T06:44:23Z")

</div>

Good afternoon scvi team! I am very new to the scvi tool and moslty have been working with scRNA-seq data within R environment. I came around a specific issue which scvi could potentially be well suited for so I would l…

[Next page](https://discourse.scverse.org/tag/scvi/10.md?match_all_tags=true&page=1&tags%5B%5D=scvi)