Archive 4 checkpoints (1-4)

paper-releases/pbisc-diffusion/updates/20261010T153319Z-manuscript-rewrite-from-deck/diffusion/dataset/server_1/manuscript/archive/pre_rewrite_20261010/tables/results_factorial.tex

@@ -0,0 +1,17 @@
+\begin{table}[t]
+\centering
+\footnotesize
+\setlength{\tabcolsep}{3pt}
+\caption{Width effects, paired within schedule part. Each entry is the mean difference over parts 6--16 ($n=11$ paired parts) with its standard deviation, and the mean divided by its standard error. Positive values mean the wider model scores higher. Composition $L_1$ is lower-better, so a positive value there is worse. Two independently trained base models resolve differences of about 0.018 in $r$ and 7 in DEG overlap (Methods).}
+\label{tab:factorial}
+\begin{tabular}{llrrr}
+\toprule
+Change & Held fixed & $\Delta r$ & $\Delta$ DEG & $\Delta$ comp. \\
+\midrule
+Trunk $1\times\to4\times$ & cond.\ 512   & $-0.014\pm0.034$ ($-1.4$) & $-11.5\pm12.9$ ($-2.9$) & $-0.000\pm0.022$ ($-0.0$) \\
+Trunk $1\times\to4\times$ & cond.\ 2,048 & $+0.003\pm0.021$ ($+0.5$) & $-4.0\pm7.7$ ($-1.7$)   & $+0.114\pm0.034$ ($+11.2$) \\
+Cond.\ $512\to2{,}048$ & trunk $1\times$ & $-0.013\pm0.018$ ($-2.3$) & $-6.8\pm7.9$ ($-2.8$)   & $-0.027\pm0.024$ ($-3.7$) \\
+Cond.\ $512\to2{,}048$ & trunk $4\times$ & $+0.004\pm0.015$ ($+0.9$) & $+0.6\pm4.8$ ($+0.4$)   & $+0.087\pm0.039$ ($+7.3$) \\
+\bottomrule
+\end{tabular}
+\end{table}

paper-releases/pbisc-diffusion/updates/20261010T153319Z-manuscript-rewrite-from-deck/diffusion/dataset/server_1/manuscript/archive/pre_rewrite_20261010/tables/results_readout.tex

@@ -0,0 +1,18 @@
+\begin{table}[t]
+\centering
+\footnotesize
+\setlength{\tabcolsep}{4pt}
+\caption{Three ways to read the support head, on the same checkpoints. Base arm, parts 14--16 of the same run, same conditions, seed and 50 deterministic steps; only the last-step rule that turns support logits into a sparsity pattern differs. The inherited default keeps a gene when its detection probability exceeds one half. The swept threshold keeps it above $\mathrm{sigmoid}(-2.0)=0.12$, the setting that maximizes $r$ in a sweep (Supplementary Note~S5). The drawn read-out samples each gene from its own probability and has no parameter. Detected genes are means per cell on the 2,980-gene panel at part 16; the observed cells have 252 (median 230). Scores are means over the three parts.}
+\label{tab:readout}
+\begin{tabular}{lrrrrr}
+\toprule
+Read-out & parameters & genes/cell & $r$ & DEG & comp.\ $L_1$ \\
+\midrule
+Cut at $p>0.5$ (inherited default) & 1 & 191 & 0.245 & 59.7 & 0.199 \\
+Cut at $p>0.12$ ($\tau=-2.0$, swept) & 1 & 286 & \textbf{0.326} & 65.7 & 0.208 \\
+Drawn from $p$ (recommended)       & \textbf{0} & 228 & 0.286 & \textbf{67.7} & \textbf{0.197} \\
+\midrule
+Observed cells & --- & 252 & --- & --- & --- \\
+\bottomrule
+\end{tabular}
+\end{table}

paper-releases/pbisc-diffusion/updates/20261010T153319Z-manuscript-rewrite-from-deck/diffusion/dataset/server_1/manuscript/archive/pre_rewrite_20261010/tables/results_realbulk.tex

@@ -0,0 +1,24 @@
+\begin{table}[t]
+\centering
+\footnotesize
+\setlength{\tabcolsep}{4pt}
+\caption{Separately measured bulk transfers the condition effect and not the donor. Cells generated from the clinical bulk of two paired cohorts (1,500 cells per sample, base arm at part 16, drawn read-out) are compared with the same donors' observed single cells. The three arms differ only in the expression vector handed to the conditioner: the assay's bulk (raw), that bulk after a ten-pair gene-wise correction fitted with the donor's batch held out (calibrated), and the donor's own pseudobulk (ceiling). Scored on the genes both assays measure in every sample. Condition effect: Pearson correlation between generated and observed within-donor $\log_2$ fold change, influenza-exposed against mock, over all genes and over the 200 genes with the largest observed change; slope is that of generated on observed. Donor recovery: fraction of samples whose summed generated cells are closest, by correlation, to their own observed pseudobulk among all samples of the cohort; chance is one over the number of samples. The last row applies the same rule to the lupus benchmark, where the model is conditioned on pseudobulk of a cohort resembling its training data.}
+\label{tab:realbulk}
+\begin{tabular}{llrrrrr}
+\toprule
+ & & \multicolumn{3}{c}{Condition effect (Randolph)} & \multicolumn{2}{c}{Donor recovery} \\
+\cmidrule(lr){3-5}\cmidrule(lr){6-7}
+Cohort & Arm & $r$, all genes & $r$, top 200 & slope & top-1 & chance \\
+\midrule
+Randolph, 89 donors $\times$ 2 & raw        & 0.635 & 0.785 & 0.355 & 0.6\%  & 0.56\% \\
+                                & calibrated & 0.644 & 0.789 & 0.439 & 1.7\%  & 0.56\% \\
+                                & pseudobulk & 0.573 & 0.736 & 0.376 & 2.8\%  & 0.56\% \\
+\addlinespace
+ImMicro, 107 donors             & raw        & --- & --- & --- & 0.9\%  & 0.93\% \\
+                                & calibrated & --- & --- & --- & 2.8\%  & 0.93\% \\
+                                & pseudobulk & --- & --- & --- & 3.7\%  & 0.93\% \\
+\addlinespace
+Lupus benchmark, 42 donors      & pseudobulk & --- & --- & --- & 31.0\% & 2.4\% \\
+\bottomrule
+\end{tabular}
+\end{table}

paper-releases/pbisc-diffusion/updates/20261010T153319Z-manuscript-rewrite-from-deck/diffusion/dataset/server_1/manuscript/archive/pre_rewrite_20261010/tables/results_sampling.tex

@@ -0,0 +1,19 @@
+\begin{table}[t]
+\centering
+\small
+\caption{Sampler steps change the effect correlation more than the seed does. One model and one condition set; only the number of deterministic DDIM steps changes ($\eta=0$, so a fixed seed gives every arm the same starting noise per cell). The last row is the spread over five sampling seeds at 50 steps, which is the resolution of a single score. Differential-gene overlap and composition are flat from 25 steps upward; $r$ falls by more than two seed standard deviations at 200 steps.}
+\label{tab:sampling}
+\begin{tabular}{lrrr}
+\toprule
+Steps & $r$ & DEG & comp.\ $L_1$ \\
+\midrule
+10  & 0.263 & 58 & 0.237 \\
+25  & 0.258 & 53 & 0.206 \\
+50  & 0.258 & 57 & 0.206 \\
+100 & 0.253 & 63 & 0.205 \\
+200 & 0.240 & 58 & 0.205 \\
+\midrule
+Seed s.d.\ at 50 steps ($n=5$) & 0.005 & 4.7 & 0.002 \\
+\bottomrule
+\end{tabular}
+\end{table}