Why scanpy Warns "adata.X seems to be already log-transformed" Even When You Never Touched X

scanpy · log1p · multi-layer normalization · false-positive warning

Why scanpy Warns "adata.X seems to be already log-transformed" Even When You Never Touched X

Status as of writing: open, reproduced against scanpy main, 2026-09. Reported in scanpy#4285. Not a data-loss bug — a misleading warning that can hide a real one.

If you work with spliced/unspliced layers (RNA velocity workflows) or any multi-layer object, you've probably run something like this and been confused by the output:

sc.pp.log1p(adata)
sc.pp.log1p(adata, layer="spliced")
sc.pp.log1p(adata, layer="unspliced")

# WARNING: adata.X seems to be already log-transformed.
# WARNING: adata.X seems to be already log-transformed.

The last two calls target layer="spliced" and layer="unspliced" — they don't touch .X at all. Yet you get the "X seems to be already log-transformed" warning twice.

The actual check

Confirmed by reading scanpy's _simple.py directly:

@log1p.register(AnnData)
def log1p_anndata(adata, *, base=None, copy=False, chunked=False,
                   chunk_size=None, layer=None, obsm=None):
    if "log1p" in adata.uns:
        logg.warning("adata.X seems to be already log-transformed.")
    ...

The check is if "log1p" in adata.uns — a single flag that gets set the first time you log1p anything on the object, and is then checked unconditionally on every subsequent call regardless of the layer argument. It never looks at the actual target layer, only at whether .X was ever log1p'd before, on any call.

Why this is more than cosmetic

Two real costs, neither of which is "the code breaks":

• Alert fatigue. If you log1p three layers per sample across dozens of samples in a loop, you get a wall of identical-looking warnings that are individually meaningless — which trains you to stop reading them. The one time this warning is actually telling you something real (you really did call log1p on .X twice — the double-normalization class of bug we've written about before), it's buried in noise.
• It's a global flag, not a per-layer one. There's no way from the warning text alone to tell which of your calls it's actually about.

What to do about it now

Since the fix isn't merged yet, the practical workaround is to not rely on the built-in warning to catch double-log1p at all — check explicitly instead:

# Before log1p-ing a layer, check its own max value directly —
# don't rely on adata.uns['log1p'] or the warning text.
import numpy as np

def already_logged(matrix, threshold=20):
    m = matrix.max() if hasattr(matrix, "max") else matrix.toarray().max()
    return m < threshold  # raw UMI counts are almost always well above this

for layer in ["spliced", "unspliced"]:
    if not already_logged(adata.layers[layer]):
        sc.pp.log1p(adata, layer=layer)

This is the same lesson as our earlier double-normalization post, applied one level up: don't trust a global "has this been transformed" signal on an object that may hold several independently-processed matrices. Check the specific array you're about to touch.

Verified directly against scanpy's main branch source, 2026-09. Source: scanpy#4285.

More preprocessing landmines

The double-normalization bug that hit our own notebooks

Pack 1 documents 7 real API-drift bugs found while re-executing all 16 notebooks — exact errors, root causes, fixes.

Get Pack 1 for $19 →

A Year-Old scanpy Bug That Could Silently Zero Out Your Data — Fixed Last Month

scanpy · normalize_total · silent data loss · CSR vs dense · preprocessing bug

A Year-Old scanpy Bug That Could Silently Zero Out Your Data — Fixed Last Month

We've written before about how easy it is to accidentally double-log-normalize spatial data (it happened across our own corpus — see our earlier workflow post). This one is a different bug in the same function, and it's worse: it can silently zero out your entire matrix with no error, just two easy-to-miss RuntimeWarnings.

Status as of writing: present in scanpy 1.11.2 through 1.12.3 (2025-05-28 → 2026-07-24). Fixed in 1.12.4 (released 2026-08-27) via PR #4256, reported in scanpy#4251. If you pinned scanpy any time in the last 14 months, check your version.

What target_sum=None is supposed to mean

sc.pp.normalize_total(adata, target_sum=None) — the default most tutorials use — documents its target as "the median of total counts for observations." Simple enough. But _normalize_total_helper computes that median two different ways depending on whether your matrix is CSR sparse or dense:

# CSR path (scanpy/preprocessing/_normalization.py, pre-1.12.4)
if target_sum is None:
    target_sum = np.median(counts_per_cell)            # median over ALL cells

# dense / everything-else path
if target_sum is None:
    target_sum = _compute_nnz_median(counts_per_cell)   # median over POSITIVE cells only

The two only disagree when a zero-total cell shifts the median — with row sums like [0, 10, 10, 20] both give 10, which is probably why this went unnoticed for so long. Empty cells are common in practice: an unfiltered raw matrix, an aggressive gene subset, a .raw slice with different filtering than the parent object.

The severe case: when at least half your cells are empty

If enough cells have zero total counts, the CSR branch's median lands on 0. Dividing by zero doesn't error — it silently produces an all-zero matrix, with only the warnings below as evidence anything went wrong:

import anndata as ad
import numpy as np
import scanpy as sc
import scipy.sparse as sp

# row sums [0, 0, 10] — the only populated cell has counts [4, 6]
dense2 = np.array([[0.0, 0.0], [0.0, 0.0], [4.0, 6.0]], dtype=np.float32)

for label, X in (("dense", dense2.copy()), ("csr", sp.csr_matrix(dense2))):
    a = ad.AnnData(X=X)
    sc.pp.normalize_total(a, target_sum=None)
    out = a.X if isinstance(a.X, np.ndarray) else a.X.toarray()
    print(f"{label:>5}: {out.ravel()}")

# dense: [0. 0.  0. 0.  4. 6.]   <- correct, target=10, populated cell untouched
#   csr: [0. 0.  0. 0.  0. 0.]   <- target=0, the ONLY real data point is destroyed

.../scanpy/preprocessing/_normalization.py:119: RuntimeWarning: divide by zero encountered in divide
.../scanpy/preprocessing/_normalization.py:119: RuntimeWarning: invalid value encountered in divide

No exception. No crash. Just a warning easy to filter out in a notebook with dozens of other warnings, and a matrix that's now all zeros.

Why the CSR path had a different rule at all

Before 1.11.2 there was a single code path, and it excluded zero-count cells on both branches. The split arrived with the numba CSR kernel in PR #3571 (merged 2025-05-19), which kept the old positive-cell-median rule for the non-CSR branch but gave the new CSR branch a plain np.median. Neither that PR nor its release notes mention a semantic change — this reads as an oversight, not a decision, and the reporter's one-line fix (route the CSR branch through the same _compute_nnz_median helper) is exactly what landed in 1.12.4.

What to actually do

• Upgrade to scanpy ≥1.12.4 if you can — this is the real fix.
• If you're pinned to an older version, pass an explicit target_sum (e.g. target_sum=1e4) instead of relying on the median default — this sidesteps the bug entirely regardless of version.
• Before normalizing, check (adata.X.sum(axis=1) == 0).sum() — if it's more than a small fraction of your cells, filter them out first rather than letting either branch guess a target around them.

Verified directly against scanpy's main branch source and the released 1.12.4 tag, 2026-09. Source: scanpy#4251.

Building a Spatial Transcriptomics Workflow: From cell2location to BANKSY and Beyond

Spatial transcriptomics · complete workflow · best practices · Squidpy · publication figures

Putting It All Together

This series has covered spatial transcriptomics from first principles (Squidpy post 1) through advanced methods — cell2location, SpatialDE2, Tangram, BANKSY, ligand-receptor analysis, trajectory analysis, and platform comparisons. This final post assembles them into a decision-driven workflow you can apply to any new Visium dataset.

The goal is not to run every tool — it is to know which tools answer which questions, in what order, and at what computational cost.

Figure 1. Left: complete spatial analysis pipeline from raw data to publication figures. Right: approximate runtime and peak memory per tool on a standard 8-core laptop with a 3,500-spot Visium dataset.

Phase 1: Data Loading and QC (30 minutes)

Every analysis starts here. Do not skip the QC plots — spatial data has tissue-specific artifacts (necrotic regions, edge effects, adipose tissue with near-zero counts) that will distort downstream results if not filtered.

import squidpy as sq
import scanpy as sc
import matplotlib.pyplot as plt

adata = sq.read.visium("spaceranger_output/", library_id="sample_01")

# QC metrics
sc.pp.calculate_qc_metrics(adata, inplace=True)
sq.pl.spatial_scatter(adata, color=["total_counts", "n_genes_by_counts"], ncols=2)

# Filter: remove low-count spots (necrosis, edge artifacts)
sc.pp.filter_cells(adata, min_counts=500)
sc.pp.filter_cells(adata, min_genes=200)
sc.pp.filter_genes(adata, min_cells=10)

# Normalize
sc.pp.normalize_total(adata, target_sum=1e4)
sc.pp.log1p(adata)
sc.pp.highly_variable_genes(adata, n_top_genes=3000, flavor="seurat_v3")

# Save raw counts for cell2location
adata.layers["counts"] = adata.X.copy()
adata.raw = adata

Phase 2: Clustering and Annotation (1–2 hours)

Build spatial graph first, then standard dimensionality reduction. The spatial graph is required for neighborhood enrichment and LR analysis later.

sq.gr.spatial_neighbors(adata, coord_type="grid", n_rings=2)

sc.tl.pca(adata, use_highly_variable=True)
sc.pp.neighbors(adata, n_neighbors=15, n_pcs=30)
sc.tl.umap(adata)
sc.tl.leiden(adata, resolution=0.3)

# Inspect: do clusters correspond to tissue morphology?
sq.pl.spatial_scatter(adata, color="leiden", size=1.5)
sc.pl.umap(adata, color="leiden")

# Annotate clusters with known markers
marker_genes = {
    "Cancer": ["EPCAM", "KRT8", "MKI67"],
    "T cells": ["CD3E", "CD8A", "CD4"],
    "Macrophages": ["CD68", "CD163", "SPP1"],
    "Fibroblasts": ["COL1A1", "FAP", "ACTA2"],
    "Endothelial": ["PECAM1", "VWF", "CDH5"],
}
sc.pl.dotplot(adata, marker_genes, groupby="leiden")

Phase 3: Choose Your Analysis Path

After annotation, your research question determines what to run next. Use this decision tree:

Q: What cell types are in each spot, at single-cell resolution?
→ cell2location (Post 12) — requires matched scRNA-seq reference

Q: Which genes show non-random spatial patterns?
→ SpatialDE2 (Post 14) or sq.gr.spatial_autocorr for a faster first pass

Q: Are cell types communicating via known ligand-receptor pairs?
→ sq.gr.nhood_enrichment + sq.gr.ligrec (Post 17)

Q: How does gene expression change across tissue regions continuously?
→ PAGA + DPT trajectory (Post 18)

Q: Are there spatial domains not captured by gene expression alone?
→ BANKSY (Post 16) — incorporates neighbor expression into clustering

Q: I have scRNA-seq from a different tissue; can I map those cells spatially?
→ Tangram (Post 15) — optimal transport mapping

Phase 4: Publication Figure Checklist

Before submitting, verify each figure against this checklist:

# 1. Spatial scatter — confirm spot size is set for the journal's figure width
sq.pl.spatial_scatter(adata, color="cell_type", size=1.2,
                       figsize=(6, 6), dpi=300, save="fig1a.pdf")

# 2. Color palettes — use perceptually uniform maps for continuous data
#    Use categorical palettes with enough contrast for 8+ cell types
import matplotlib.pyplot as plt
# Bad: plt.cm.jet (never use in publications)
# Good: "plasma", "viridis", "RdBu_r" for continuous; "tab20" for categories

# 3. Scale bar — always include physical scale
# sq.pl.spatial_scatter adds it automatically from adata.uns["spatial"] metadata

# 4. Export at 300 dpi minimum for print
plt.savefig("figure1.pdf", dpi=300, bbox_inches="tight")
plt.savefig("figure1.png", dpi=300, bbox_inches="tight")

Compute Budget for a Full Analysis

From the benchmark figure, planning a full Visium TME analysis on a laptop (8 cores, 16 GB RAM):

• QC + clustering: ~30 min, <2 GB RAM
• cell2location (1000 epochs): ~45 min, 6 GB RAM — the bottleneck; use GPU if available
• SpatialDE2: ~8 min, 3.5 GB RAM
• sq.gr.ligrec (1000 perms): ~12 min, 2 GB RAM
• PAGA + DPT: ~5 min, 1.5 GB RAM
• Total: ~1.5–2 hours end-to-end

If you are running on a server with a GPU, cell2location drops to under 10 minutes and the full pipeline completes in well under an hour.

What This Series Has Not Covered

Spatial transcriptomics is a fast-moving field. Topics worth following as they mature:

• 3D spatial reconstruction — stacking serial sections with PASTE2 or Spateo
• Spatial proteomics integration — combining Visium with CODEX or IMC protein data
• Spatial multi-sample integration — harmonizing multiple patients/conditions with Harmony or scVI adapted for spatial
• Foundation models — Geneformer and scGPT adapted for spatial context are emerging

All code from this series is available on GitHub. The Squidpy Starter Pack ($19) contains runnable Jupyter notebooks for posts 1–5 with a real Visium PBMC dataset. For the immunotherapy biology behind the cell types you are mapping, see the sister blog NeoantigenLab.

Visium HD: What Changes When Resolution Goes Subcellular

Spatial transcriptomics · Visium HD · Xenium · CosMx · platform comparison

Why Resolution Matters (and Why It Is Not Everything)

Visium HD delivers 2 µm bins — roughly single-cell resolution — while the original Visium captures 55 µm spots that each contain 5–20 cells. That 27x improvement in resolution sounds decisive. In practice, the right platform depends on your biological question, not just the spec sheet.

This post covers what Visium HD actually changes analytically, where it outperforms its predecessor, and when Xenium or CosMx is the better choice despite having a fixed gene panel.

Figure 1. Left: spatial resolution comparison — Visium HD achieves 2 µm bins vs. 55 µm spots for Visium v1. Right: capability trade-offs across platforms normalized to maximum observed values.

What Visium HD Changes: The Bin Model

Visium HD does not capture individual cells — it captures 2 µm square bins that tile the tissue continuously. The key workflow difference is binning strategy: you choose a bin size at analysis time, not at capture time.

import squidpy as sq
import scanpy as sc

# Load Visium HD at 8 µm bins (good starting point — balances resolution and sparsity)
adata = sq.read.visium(
    path="visium_hd_output/",
    counts_file="filtered_feature_bc_matrix.h5",
    library_id="your_sample_id",
    bin_size=8,  # options: 2, 8, 16 micrometers
)

print(adata)
# AnnData: n_obs (bins) depends heavily on bin_size
# bin_size=2:  ~500,000 bins (very sparse, ~2-5 UMI/bin)
# bin_size=8:  ~30,000 bins (workable, ~50-100 UMI/bin)
# bin_size=16: ~8,000 bins (Visium-like density)

Start with 8 µm bins for exploratory analysis. At 2 µm, most bins will have fewer than 5 UMI — too sparse for reliable clustering without heavy smoothing. Only use 2 µm bins if you need subcellular resolution for a specific question (e.g., resolving nuclear vs. cytoplasmic transcript localization).

The Critical Difference: Deconvolution Is Optional

With Visium v1, cell2location or RCTD is mandatory to get cell type proportions per spot. With Visium HD at 8 µm bins, many bins will contain a single dominant cell type and can be annotated directly by marker expression — though mixed bins still exist at tissue interfaces.

# Standard clustering works directly on HD bins
sc.pp.filter_cells(adata, min_counts=10)
sc.pp.normalize_total(adata, target_sum=1e4)
sc.pp.log1p(adata)
sc.pp.highly_variable_genes(adata, n_top_genes=3000)
sc.tl.pca(adata)
sc.pp.neighbors(adata, n_neighbors=15)
sc.tl.leiden(adata, resolution=0.4)

# Annotate by marker expression (same workflow as scRNA-seq)
sc.pl.dotplot(adata, markers_dict, groupby="leiden")

# Optional: if you want single-cell precision at 2µm bins,
# still use cell2location with matched scRNA-seq reference

Visium HD vs. Xenium: Choosing the Right Tool

Choose Visium HD when:

• You need whole-transcriptome coverage (not a fixed panel)
• Your question involves discovery — finding unexpected cell states or pathways
• You have FFPE tissue (Visium HD supports FFPE; Xenium also does)
• You want spatial multi-omics (Visium HD + CUT&TAG or protein co-detection)

Choose Xenium/CosMx when:

• You need subcellular resolution (mRNA localization within a cell)
• Your target genes are already known and a panel of 100–500 genes is sufficient
• You need higher sensitivity per transcript (in-situ amplification vs. sequencing)
• Your tissue is thick or 3D (Xenium handles z-stacking)

Squidpy Compatibility

Visium HD AnnData loads exactly like Visium v1 from Squidpy's perspective. All tools covered in this series work without modification:

# Everything from earlier posts works identically on Visium HD
sq.gr.spatial_neighbors(adata, coord_type="generic", n_neighs=6)
sq.gr.nhood_enrichment(adata, cluster_key="leiden")
sq.gr.ligrec(adata, n_perms=500, cluster_key="leiden")
sq.gr.spatial_autocorr(adata, mode="moran")

# The only practical difference: more bins = longer runtime
# nhood_enrichment on 30,000 bins takes ~10 min vs ~2 min for Visium v1

One Practical Warning: Tissue Coverage

Visium HD captures a 6.5 mm × 6.5 mm area — identical to Visium v1. This is enough for most tissue sections but smaller than a typical whole-slide image. If your question requires comparing multiple tumor regions across a large section, you will need multiple capture areas or switch to a platform with larger footprint (some CosMx runs cover full slides).

Interested in what the immune cell types you are mapping at single-cell resolution are actually doing in the tumor? NeoantigenLab covers T cell exhaustion, checkpoint inhibitor mechanisms, and CAR-T biology at the molecular level.

Get the complete pack

SpatiaBio Squidpy Pack 1 — 16 Notebooks ($19)

All 16 notebooks — neighborhood enrichment, co-occurrence, ligand-receptor, spatial autocorrelation, publication figures, and the complete workflow. Includes memory tips for Visium HD. Ready to run on your own data.

Get it for $19 →

---

From the NeoantigenLab sister blog

Working with tumor transcriptomics for neoantigen research?

NeoantigenLab covers neoantigen biology, WES pipelines, pVACseq, and HLA typing for experimental researchers — the biology behind what you're sequencing.

Visit NeoantigenLab →

Spatial Trajectory Analysis: PAGA and Diffusion Pseudotime vs. Spateo

Spatial transcriptomics · trajectory analysis · pseudotime · Squidpy · Spateo

The Problem: Clustering Loses Spatial Continuity

Standard spatial analysis assigns each spot to a discrete cluster. But tissue biology is continuous — there is no hard boundary between stroma and tumor, between hypoxic core and normoxic periphery, between vascularized and avascular zones. Trajectory analysis reconstructs this continuum by ordering spots along a path through gene expression space, anchored to their physical location in tissue.

The result is pseudotime: a one-dimensional coordinate that captures how far each spot is along a biological transition. Paired with spatial coordinates, it reveals not just that gene expression changes, but exactly where in the tissue those changes happen.

Figure 1. Left: spatial trajectory colored by pseudotime from stroma (early) to tumor core (late). Right: gene expression dynamics along the trajectory — ECM genes decline, proliferation and immune markers peak at different positions.

Two Approaches: PAGA + Diffusion Pseudotime vs. Spateo

You have two practical options for spatial trajectory analysis:

PAGA + DPT (Scanpy/Squidpy) — Works on the spatial graph. Mature, well-documented. Best when you already have a spatial graph and want trajectories between defined regions. Requires a root cell/spot (the biological starting point).

Spateo — Purpose-built for spatial trajectories. Handles 3D spatial data and multiple trajectory branches. More flexible root selection. Heavier dependency chain.

For most Visium workflows, start with PAGA + DPT — it integrates cleanly with your existing Squidpy AnnData.

Running PAGA + DPT on Visium Data

import scanpy as sc
import squidpy as sq
import numpy as np

# Start from annotated Visium AnnData
adata = sc.read_h5ad("visium_annotated.h5ad")

# 1. Build neighborhood graph on HVGs (not spatial — for PAGA topology)
sc.pp.neighbors(adata, n_neighbors=15, use_rep="X_pca")

# 2. Run PAGA to get cluster connectivity
sc.tl.paga(adata, groups="spatial_domain")
sc.pl.paga(adata, threshold=0.1, node_size_scale=2)

# 3. Initialize UMAP from PAGA for better trajectory layout
sc.tl.umap(adata, init_pos="paga")

# 4. Set root: spot with highest stroma marker expression
stroma_score = adata[:, "COL1A1"].X.toarray().flatten()
adata.uns["iroot"] = int(np.argmax(stroma_score))

# 5. Run diffusion pseudotime
sc.tl.dpt(adata)
# Result: adata.obs["dpt_pseudotime"]

Spatial Visualization of Pseudotime

Once pseudotime is computed, project it back onto tissue coordinates:

sq.pl.spatial_scatter(
    adata,
    color="dpt_pseudotime",
    cmap="plasma",
    size=1.5,
    title="Pseudotime — stroma to tumor core",
)

# Overlay with cell type to validate the trajectory makes biological sense
sq.pl.spatial_scatter(
    adata,
    color=["dpt_pseudotime", "spatial_domain"],
    ncols=2,
    size=1.5,
)

If pseudotime increases smoothly from stroma toward tumor core, the trajectory is capturing real biology. If it looks noisy or inverted, try a different root spot — pseudotime direction is set by your root choice, not by the algorithm.

Finding Genes That Change Along the Trajectory

The most useful output is the list of genes whose expression correlates with pseudotime — these are candidates for signals driving the stroma-to-tumor transition:

from scipy.stats import spearmanr
import pandas as pd

pt = adata.obs["dpt_pseudotime"].values
gene_matrix = adata.X.toarray()  # or adata.layers["normalized"]

correlations = []
for i, gene in enumerate(adata.var_names):
    rho, pval = spearmanr(pt, gene_matrix[:, i])
    correlations.append({"gene": gene, "rho": rho, "pval": pval})

corr_df = pd.DataFrame(correlations)
corr_df["qval"] = corr_df["pval"] * len(corr_df)  # Bonferroni
corr_df = corr_df[corr_df["qval"] < 0.05].sort_values("rho", ascending=False)

print("Top increasing along trajectory:")
print(corr_df.head(10)[["gene", "rho"]])

print("
Top decreasing:")
print(corr_df.tail(10)[["gene", "rho"]])

Interpreting the Results

From the figure, four biologically coherent patterns emerge along the stroma-to-tumor trajectory:

• COL1A1 (ECM marker): Highest in stroma, decays toward tumor core — validates that the trajectory direction is correct
• CD8A (cytotoxic T cells): Peaks at the interface zone, consistent with T cell exclusion from the tumor core by immunosuppressive signals
• CXCL10 (interferon response): Transient peak at the interface — the zone of active immune engagement
• MKI67 (proliferation): Rises sharply in the tumor core, confirming that high-pseudotime spots correspond to actively dividing cancer cells

Combining with Ligand-Receptor Analysis

Trajectory analysis and LR analysis are complementary. Once you know which spatial zones correspond to early vs. late pseudotime, you can ask: which ligand-receptor interactions are active specifically at the interface? Filter spots by pseudotime bin before running gr.ligrec:

# Subset to interface zone (pseudotime 0.4–0.6)
interface_mask = (adata.obs["dpt_pseudotime"] > 0.4) & (adata.obs["dpt_pseudotime"] < 0.6)
adata_interface = adata[interface_mask].copy()

sq.gr.spatial_neighbors(adata_interface, coord_type="generic", n_neighs=6)
sq.gr.ligrec(adata_interface, n_perms=500, cluster_key="cell_type")

This reveals whether the CD8A peak and CXCL10 burst at the interface are accompanied by specific T cell - cancer cell interactions that might be candidates for checkpoint blockade potentiation.

For the immunotherapy biology behind the interface dynamics — why T cells stall at the tumor boundary and what checkpoint inhibitors do about it — see NeoantigenLab.

---

From the NeoantigenLab sister blog

Working with tumor transcriptomics for neoantigen research?

NeoantigenLab covers neoantigen biology, WES pipelines, pVACseq, and HLA typing for experimental researchers — the biology behind what you're sequencing.

Visit NeoantigenLab →