[Seurat v5] IntegrateData error: model.list[[which(reference.model)]] : subscript out of bounds

[QianhuiWan]

I get this error every time, no matter how I filter features in my input object list for IntegrateData(). Could anyone help me figure it out? Thank you so much!

# 3) feature selection + pre SCT
features <- 
  SelectIntegrationFeatures(object.list = obj_all_fil_splitBy_batches_SCT, 
                            assay=rep("slide_SCT", 
                                      length(obj_all_fil_splitBy_batches_SCT)),
                            nfeatures = 3000)
# intersect feature names between batches 
# features <- intersect(features, Reduce(intersect, lapply(obj_all_fil_splitBy_batches_SCT, \(o) rownames(o[["SCT"]]))))

obj_all_fil_splitBy_batches_SCT_preINT <- 
  PrepSCTIntegration(object.list = obj_all_fil_splitBy_batches_SCT, 
                     assay=rep("slide_SCT", 
                               length(obj_all_fil_splitBy_batches_SCT)),
                     anchor.features = features, verbose = TRUE)

## use CCA method to anchor, and reference = batch1/slide1
anchors <- FindIntegrationAnchors(
  object.list = obj_all_fil_splitBy_batches_SCT_preINT,
  assay=rep("slide_SCT", 
            length(obj_all_fil_splitBy_batches_SCT_preINT)),
  normalization.method = "SCT",
  anchor.features = features,
  reduction = "cca",     
  # batch1/slide1 as reference
  reference = 1,
  # --pick_anchors 5
  k.anchor = 5,     
  # --filter_anchors 200
  k.filter = 200, 
  # --score_anchors 30
  k.score = 30,   
  # --l2_norm
  l2.norm = TRUE,     
  # Nearest neighbor finding method: use rann not annoy
  nn.method = "rann",
  verbose = TRUE
)

## integrate other 2 batches to reference
obj_all_fil_splitBy_batches_SCT_slideINT <- IntegrateData(
  anchorset = anchors, 
  normalization.method = "SCT",
  new.assay.name = "SCT_slideINT",
  preserve.order = TRUE,
  verbose = TRUE)
[1] 1
Warning: Layer counts isn't present in the assay object; returning NULL
[1] 2
Warning: Layer counts isn't present in the assay object; returning NULL
[1] 3
Warning: Layer counts isn't present in the assay object; returning NULL
Warning: Layer counts isn't present in the assay object; returning NULL
Error in model.list[[which(reference.model)]] : subscript out of bounds
In addition: Warning messages:
1: Different cells and/or features from existing assay slide_SCT 
2: Different cells and/or features from existing assay slide_SCT 
3: Different cells and/or features from existing assay slide_SCT 

[QianhuiWan]

Hi, it’s working now. I joined the sample layers before the SCT step, and that fixed it.