Sync from GitHub via hub-sync

README.md

@@ -1,14 +1,433 @@
 ---
-title: Bartab
-emoji: 📉
-colorFrom: green
-colorTo: purple
+title: bartab
+emoji: 🍹
+colorFrom: blue
+colorTo: green
 sdk: gradio
-sdk_version: 6.13.0
+sdk_version: "5.50.0"
 app_file: app.py
-pinned: false
-license: mit
-short_description: Fitness from sequencing-based pooled competition experiments
+pinned: true
+short_description: Fitness from pooled competition experiments.
+tags:
+  - biology
+  - sequencing
+  - pooled-screen
+  - fitness
+  - gradio
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# 🍹 bartab
+
+**Estimate strain fitness from sequencing-based pooled competition experiments.**
+
+`bartab` analyses pooled barcode sequencing experiments in which multiple strains,
+mutants, guides, or constructs are grown together and quantified by NGS over time.
+It estimates the relative fitness of each barcode compared with a reference strain,
+typically WT.
+
+[![Open in Spaces](https://huggingface.co/datasets/huggingface/badges/resolve/main/open-in-hf-spaces-md-dark.svg)](https://huggingface.co/spaces/scbirlab/bartab)
+
+---
+
+## What this app does
+
+Upload three tables:
+
+1. **Count table**  
+   Read or UMI counts for each barcode in each sequencing sample.
+
+2. **Sample sheet**  
+   Metadata describing each sample: sample ID, culture/replicate ID, timepoint,
+   optional concentration, optional growth measurement, and optional sampled volume.
+
+3. **Barcode sheet**  
+   Barcode or strain identifiers, plus optional strain-level metadata.
+
+`bartab` then estimates how each barcode changes relative to a reference barcode as
+the culture expands.
+
+---
+
+### Option 1: Spike-in normalisation
+
+Use this when your experiment includes a non-growing spike-in barcode, such as a
+heat-killed strain, plasmid-only control, or other fitness-zero control.
+
+If the spike-in does not grow, then changes in the spike-in:WT count ratio report how
+much WT has expanded.
+
+For a non-growing spike-in:
+
+$$
+\log \left(
+\frac{c_{spike}(t)}{c_{wt}(t)}
+\frac{c_{wt}(0)}{c_{spike}(0)}
+\right)
+=
+-
+\log
+\frac{n_{wt}(t)}{n_{wt}(0)}
+$$
+
+Substituting this into the barcode model gives:
+
+$$
+\log \left(
+\frac{c_i(t)}{c_{wt}(t)}
+\frac{c_{wt}(0)}{c_i(0)}
+\right)
+=
+\left(
+1 - \frac{w_i}{w_{wt}}
+\right)
+\log \left(
+\frac{c_{spike}(t)}{c_{wt}(t)}
+\frac{c_{wt}(0)}{c_{spike}(0)}
+\right)
+$$
+
+So, when using spike-in normalisation, `bartab` estimates relative fitness from the
+relationship between:
+
+- the barcode:WT log-ratio change
+- the spike-in:WT log-ratio change
+
+---
+
+### Option 2: Growth-based normalisation
+
+Use this when you have measured culture growth directly, for example:
+
+- OD600
+- CFU/mL
+- estimated generations
+- another density-like growth measurement
+
+In this mode, `bartab` uses the supplied growth column as the culture expansion axis
+instead of estimating expansion from a spike-in barcode.
+
+---
+
+## Analysis modes
+
+### 👟 Fitness in a single condition
+
+This is the standard mode for pooled competition assays.
+
+`bartab` fits a weighted least-squares model to estimate relative fitness for each
+barcode.
+
+Main outputs:
+
+| Column | Meaning |
+|---|---|
+| `fitness` | relative fitness compared with the reference barcode |
+| `fitness_low` | lower confidence interval bound |
+| `fitness_high` | upper confidence interval bound |
+| `slope_p` | p-value for deviation from neutral fitness |
+
+Interpretation:
+
+| Fitness | Interpretation |
+|---|---|
+| `1` | grows like the reference |
+| `< 1` | growth disadvantage |
+| `> 1` | growth advantage |
+
+---
+
+### 📉 Dose response
+
+Use this when the sample sheet contains a concentration column.
+
+`bartab` first estimates barcode fitness across concentrations, then fits a
+two-parameter Hill model to estimate dose-response parameters.
+
+Main outputs:
+
+| Column | Meaning |
+|---|---|
+| `log_ic50` | log10 concentration giving 50% inhibition |
+| `log_ic50_p` | p-value associated with the IC50 estimate |
+
+Lower IC50 values indicate greater sensitivity to the inducer or drug.
+
+---
+
+## Required input tables
+
+The app lets you select the relevant columns after upload.
+
+Supported file formats:
+
+- `.csv`
+- `.tsv`
+- `.txt`
+- `.xlsx`
+
+---
+
+### 1. Count table
+
+One row per barcode per sample.
+
+Required columns:
+
+| Meaning | Example |
+|---|---|
+| barcode / strain identifier | `strain_id` |
+| sample identifier | `sample_id` |
+| read or UMI count | `count` |
+
+Example:
+
+| strain_id | sample_id | count |
+|---|---:|---:|
+| wt | sample_0 | 12034 |
+| mutant_A | sample_0 | 8312 |
+| spike | sample_0 | 5021 |
+| wt | sample_1 | 18420 |
+| mutant_A | sample_1 | 6420 |
+| spike | sample_1 | 2100 |
+
+---
+
+### 2. Barcode sheet
+
+One row per barcode.
+
+Required columns:
+
+| Meaning | Example |
+|---|---|
+| barcode / strain identifier | `strain_id` |
+
+Optional metadata columns are carried through to the output.
+
+Example:
+
+| strain_id | gene | annotation |
+|---|---|---|
+| wt | WT | reference |
+| mutant_A | geneA | deletion mutant |
+| spike | spike | non-growing spike-in |
+
+---
+
+### 3. Sample sheet
+
+One row per sequencing sample.
+
+Required columns:
+
+| Meaning | Example |
+|---|---|
+| sample identifier | `sample_id` |
+| culture / biological replicate | `replicate` |
+| timepoint | `timepoint` |
+
+Optional columns:
+
+| Meaning | Example | Used for |
+|---|---|---|
+| concentration | `dose` | dose-response analysis |
+| growth measurement | `growth` | growth-based normalisation |
+| sampled volume | `volume` | adaptive-volume sampling |
+
+Example:
+
+| sample_id | replicate | timepoint | dose | growth | volume |
+|---|---|---:|---:|---:|---:|
+| sample_0 | rep1 | 0 | 0 | 0.05 | 1.0 |
+| sample_1 | rep1 | 1 | 0 | 0.20 | 1.0 |
+| sample_2 | rep1 | 2 | 0 | 0.80 | 1.0 |
+
+---
+
+## Example datasets
+
+The Space includes synthetic example datasets for:
+
+- single-concentration analysis using spike-in normalisation
+- single-concentration analysis using growth-based normalisation
+- dose-response analysis using spike-in normalisation
+- dose-response analysis using growth-based normalisation
+
+Start with these examples if you want to see the expected table structure.
+
+---
+
+## Outputs
+
+After analysis, the app returns:
+
+1. **Fitted parameter table**  
+   A CSV file containing estimated fitness or dose-response parameters.
+
+2. **Annotated `.h5ad` object**  
+   A full `AnnData` object containing input data, metadata, transformations,
+   fitted values, and model outputs.
+
+3. **Diagnostic plots**
+
+Depending on the analysis mode, plots include:
+
+- time vs count
+- expansion vs count
+- time vs ratio
+- expansion vs ratio
+- predicted vs observed
+- volcano plot
+- dose-response curves
+
+---
+
+## Practical guidance
+
+- Use consistent barcode and sample identifiers across all three tables.
+- If using spike-in normalisation, include the spike-in barcode in both the count table
+and barcode sheet.
+- If using growth-based normalisation, include a numeric growth column in the sample sheet.
+- For dose-response analysis, concentration values must be numeric.
+- Low-count barcodes may give unstable estimates.
+- Always inspect diagnostic plots before interpreting individual hits.
+
+---
+
+## The model
+
+[Interactive tutorial on analysis principles](https://huggingface.co/spaces/scbirlab/tutorial-seq-fitness)
+
+In a pooled growth experiment, strains compete in the same culture. Their absolute
+growth curves may be complex because the pool eventually approaches carrying capacity.
+However, if we compare every strain to a reference strain, the shared density-dependent
+term cancels. The reference strain’s expansion can then be used as the effective
+growth clock.
+
+For strain $i$ relative to WT:
+
+$$
+\log n_i(t)
+=
+\frac{w_i}{w_{wt}}
+\log \frac{n_{wt}(t)}{n_{wt}(0)}
++
+\log n_i(0)
+$$
+
+where:
+
+- $n_i(t)$ is the abundance of strain $i$ at time $t$
+- $w_i$ is the intrinsic growth rate of strain $i$
+- $w_i / w_{wt}$ is the relative fitness
+
+So the problem becomes: estimate how barcode abundance changes relative to WT as WT
+expands.
+
+---
+
+## From cells to sequencing counts
+
+In practice, we do not observe cell numbers directly. We observe read counts:
+
+$$
+c_i(t)
+$$
+
+These are affected by sampling, library preparation, and sequencing depth. To remove
+sample-specific sequencing depth effects, `bartab` works with ratios of barcode counts
+to the reference barcode.
+
+The key quantity is:
+
+$$
+\log \left(
+\frac{c_i(t)}{c_{wt}(t)}
+\frac{c_{wt}(0)}{c_i(0)}
+\right)
+$$
+
+This is the log-change in the abundance of barcode $i$ relative to WT, normalised
+to the starting timepoint.
+
+Under the model:
+
+$$
+\log \left(
+\frac{c_i(t)}{c_{wt}(t)}
+\frac{c_{wt}(0)}{c_i(0)}
+\right)
+=
+\left(
+\frac{w_i}{w_{wt}} - 1
+\right)
+\log
+\frac{n_{wt}(t)}{n_{wt}(0)}
+$$
+
+Thus, each barcode should follow an approximately straight line. The slope gives the
+barcode’s fitness relative to WT.
+
+---
+
+## Estimating culture expansion
+
+The remaining problem is that the true WT expansion,
+
+$$
+\frac{n_{wt}(t)}{n_{wt}(0)}
+$$
+
+is usually not directly observed. `bartab` supports two ways to estimate it.
+
+---
+
+## When this model is appropriate
+
+`bartab` is designed for pooled competition experiments where:
+
+- barcodes identify strains, mutants, guides, or constructs
+- all barcodes are grown together in shared cultures
+- barcode abundance is quantified by sequencing
+- one barcode can be treated as a reference
+- expansion can be estimated from a spike-in or measured growth
+
+It is especially useful for:
+
+- bacterial pooled competition assays
+- barcoded mutant libraries
+- CRISPRi or guide-based growth assays
+- chemical-genetic pooled fitness experiments
+- inducer or drug dose-response screens
+
+---
+
+## Limitations
+
+`bartab` estimates relative fitness, not absolute growth rate.
+
+Results may be unreliable when:
+
+- the reference barcode is depleted or poorly counted
+- the spike-in is not truly non-growing
+- barcode counts are extremely low
+- barcode identities are mismatched between tables
+- bottlenecks dominate the experiment
+- strong barcode-specific sequencing biases are present
+- timepoints or concentrations are too sparse for model fitting
+
+For dose-response fitting, IC50 estimates are most meaningful when the tested
+concentration range brackets the transition from weak to strong inhibition.
+
+---
+
+## Local use
+
+To run the app locally:
+
+```bash
+pip install -r requirements.txt
+gradio app.py
+```
+
+For package and source code, see https://github.com/scbirlab/bartab

app.py

@@ -0,0 +1,841 @@
+"""Gradio demo for bartab."""
+
+from typing import Iterable, List, Union
+from functools import partial
+from io import TextIOWrapper
+import os
+os.environ["COMMANDLINE_ARGS"] = "--no-gradio-queue"
+
+from carabiner import cast, print_err
+from carabiner.decorators import decorator_with_params
+from carabiner.pd import read_table
+import gradio as gr
+import nemony as nm
+import numpy as np
+import pandas as pd
+
+import anndata
+anndata.settings.allow_write_nullable_strings = True
+import bartab
+from bartab.io import load_anndata
+from bartab.models.anndata import AnnDataWLSModel, AnnDataHillModel
+from bartab.plotting import (
+    dose_response,
+    expansion_vs_count,
+    expansion_vs_ratio, 
+    pred_vs_true,
+    time_vs_count, 
+    time_vs_ratio, 
+    volcano
+)
+from bartab.transforms import compute_log_ratios
+
+pd.options.future.infer_string = True
+
+MODES: dict = {
+    "single": "👟 Fitness in a single condition", 
+    "dose response": "📉 Fitness dose response to CRISPRi inducer",
+}
+def _message(s: str):
+    print_err(f"[INFO] {s}")
+    gr.Info(s, duration=10)
+    return None
+
+
+def load_input_data(
+    filename: str, 
+    cols: Iterable
+) -> List[pd.DataFrame]:
+    df = read_table(filename)
+    print_err(df)
+    out = [gr.update(value=df, visible=False)]
+    for key, col in cols.items():
+        if isinstance(col, tuple):
+            col_type = col[1]
+            if col_type == "string":
+                choices = list(df.select_dtypes(include="str"))
+            elif col_type == "numeric":
+                choices = list(df.select_dtypes(include="number"))
+            else:
+                choices = list(df)
+        else:
+            choices = list(df)
+        choices = [""] + choices
+        print_err(key, f"{choices=}")
+        out.append(
+            gr.update(
+                choices=choices,
+                value=key if key in choices else choices[0],
+                interactive=True, 
+                visible=True,
+            )
+        )
+    print_err(out)
+    return out
+
+
+def load_barcode_names(
+    df: pd.DataFrame,
+    strain_col: str
+) -> List[List[str]]:
+    strains = sorted(df[strain_col].unique())
+    print_err(strain_col, f"{strains=}")
+    return gr.update(
+        choices=strains,
+        value="wt" if "wt" in strains else strains[0],
+        interactive=True, 
+        visible=True,
+    ), gr.update(
+        choices=strains, 
+        value="spike" if "spike" in strains else "",
+        interactive=True, 
+        visible=True,
+    ), gr.update(
+        choices=strains, 
+        value=[],
+        allow_custom_value=True,
+        interactive=True, 
+        visible=True,
+    )
+    
+
+def _prepare_to_fit(
+    counts: pd.DataFrame,
+    strain_sheet: pd.DataFrame,
+    sample_sheet: pd.DataFrame,
+    count_column: str,
+    strain_id_column: str,
+    timepoint_column: str,
+    concentration_column: str,
+    sample_id_column: str,
+    culture_id_column: str,
+    volume_column: str = "volume",
+    growth_column: str = "growth",
+    reference: str = "wt",
+    spike_name: str = "spike",
+    spike_mode: str = "Spike",
+    growth_type: str = "density",
+    pseudocount: float = 1.
+):
+    use_spike = (spike_mode == "Spike")
+    adata = load_anndata(
+        counts=counts,
+        sample_meta=sample_sheet,
+        strain_meta=strain_sheet,
+        reference=reference,
+        count_column=count_column,
+        timepoint_column=timepoint_column,
+        # t0=args.t0,
+        concentration_column=concentration_column,
+        strain_id=strain_id_column,
+        sample_id=sample_id_column,
+        culture_id=culture_id_column,
+        spike=spike_name if spike_name else None,
+    )
+    print_err(adata)
+    adata = compute_log_ratios(
+        adata=adata,
+        pseudocount=pseudocount,
+        volume_column=volume_column,
+        growth_column=growth_column,
+        growth_type=growth_type,
+        use_spike=use_spike,
+    )
+    print_err(adata)
+    return adata
+
+
+def do_analysis(*args):
+    args = [
+        a if not (isinstance(a, str) and a == "") else None 
+        for a in args
+    ]
+    mode = args[-1]
+    concentration_column = args[3]
+    print_err(args[:-1])
+    try:
+        adata = _prepare_to_fit(*args[:-1])
+    except TypeError as e:
+        print_err(*args[:-1])
+        raise e
+    _message("Using a weighted least squares model")
+    model = AnnDataWLSModel()
+    results = model.fit(adata=adata)
+    if mode == MODES["dose response"]:
+        _message(
+            "Using a Hill non-linear model with "
+            f"'{concentration_column}' for concentration."
+        )
+        model = AnnDataHillModel()
+        results = model.fit(adata=results, concentration=concentration_column)
+    return gr.update(value=results.obs, visible=True), results
+
+
+def _fig2img(fig):
+    import PIL
+    # img = PIL.Image.frombytes(
+    #     "RGBa", 
+    #     fig.canvas.get_width_height(),
+    #     fig.canvas.buffer_rgba(),
+    # )
+    import io
+    buf = io.BytesIO()
+    fig.savefig(buf)
+    buf.seek(0)
+    img = PIL.Image.open(buf)
+    return img
+
+
+@decorator_with_params
+def _plot_wrapper(fn, message="Plotting..."):
+    def _fn(*args, **kwargs):
+        if args[1] == "":
+            args[1] = None
+        if message:
+            _message(message)
+        fig, axes = fn(*args, **kwargs)
+        if isinstance(fig, tuple) and isinstance(axes, bool):
+            fig, vis = fig
+            return gr.update(
+                value=_fig2img(fig) if fig is not None else fig, 
+                visible=vis,
+            )
+        else:
+            return gr.update(value=_fig2img(fig), visible=True)
+    return _fn
+
+
+@_plot_wrapper()
+def _plot_dose_response(
+    adata,
+    highlight=None,
+    control_prefix: str = "ctrl_",
+    mode: str = MODES["single"]
+):
+    do_dose_response = mode == MODES["dose response"]
+    if do_dose_response:
+        print_err("Plotting dose response")
+        fig, axes = dose_response(
+            adata,
+            highlight_barcodes=highlight,
+            model_name="WLS",
+            control_prefix=control_prefix,
+        )
+        return fig, axes
+    else:
+        print_err("Skipping dose response")
+        return (None, None), False
+
+
+@_plot_wrapper(message="Plotting time vs count")
+def _plot_time_vs_count(
+    adata,
+    highlight=None,
+    control_prefix: str = "ctrl_",
+    *args, **kwargs
+):
+    return time_vs_count(
+        adata,
+        highlight_barcodes=highlight,
+        control_prefix=control_prefix,
+    )
+
+
+@_plot_wrapper(message="Plotting expansion vs count")
+def _plot_expansion_vs_count(
+    adata,
+    highlight=None,
+    control_prefix: str = "ctrl_",
+    *args, **kwargs
+):
+    return expansion_vs_count(
+        adata,
+        highlight_barcodes=highlight,
+        control_prefix=control_prefix,
+    )
+
+
+@_plot_wrapper(message="Plotting time vs ratio")
+def _plot_time_vs_ratio(
+    adata,
+    highlight=None,
+    control_prefix: str = "ctrl_",
+    *args, **kwargs
+):
+    return time_vs_ratio(
+        adata,
+        highlight_barcodes=highlight,
+        control_prefix=control_prefix,
+    )
+
+
+@_plot_wrapper(message="Plotting expansion vs ratio")
+def _plot_expansion_vs_ratio(
+    adata,
+    highlight=None,
+    control_prefix: str = "ctrl_",
+    *args, **kwargs
+):
+    return expansion_vs_ratio(
+        adata,
+        highlight_barcodes=highlight,
+        control_prefix=control_prefix,
+    )
+
+@_plot_wrapper(message="Plotting predicted vs observed")
+def _plot_pred_vs_true(
+    adata,
+    highlight=None,
+    control_prefix: str = "ctrl_",
+    mode: str = MODES["single"]
+):
+    do_dose_response = mode == MODES["dose response"]
+    return pred_vs_true(
+        adata,
+        model_name="HillFitnessModel" if do_dose_response else "WLS",
+        highlight_barcodes=highlight,
+        control_prefix=control_prefix,
+    )
+
+
+@_plot_wrapper(message="Plotting volcano")
+def _plot_volcano(
+    adata,
+    highlight=None,
+    control_prefix: str = "ctrl_",
+    mode: str = MODES["single"]
+):
+    do_dose_response = mode == MODES["dose response"]
+    return volcano(
+        adata,
+        highlight_barcodes=highlight,
+        control_prefix=control_prefix,
+        model_name="HillFitnessModel" if do_dose_response else "WLS",
+        param="ic50" if do_dose_response else "fitness",
+        xscale="log" if do_dose_response else "linear",
+        vline=None if do_dose_response else 1.,
+        p="log_ic50_p" if do_dose_response else "slope_p",
+    )
+
+
+def download_tables(
+    df: pd.DataFrame,
+    adata
+) -> str:
+    df_hash = nm.hash(pd.util.hash_pandas_object(df).values)
+    filename = f"bartab-{df_hash}"
+    filename_csv = f"{filename}.csv"
+    df.to_csv(filename, index=False)
+    filename_adata = f"{filename}.h5ad"
+    adata.write(filename_adata)
+    return gr.update(
+        value=filename_csv, 
+        visible=True,
+    ), gr.update(
+        value=filename_adata, 
+        visible=True,
+    )
+
+
+
+def _file_input(**kwargs):
+    return partial(gr.File,
+        file_types=[".xlsx", ".csv", ".tsv", ".txt"],
+    )(**kwargs)
+
+
+def _load_from_file(*args):
+    if len(args) > 1:
+        return args
+    else:
+        return args[0]
+
+
+def _invisible_dropdown(**kwargs):
+    return partial(gr.Dropdown,
+        choices=[],
+        interactive=False,
+        visible=True,
+    )(**kwargs)
+
+
+def _invisible_plot(**kwargs):
+    return partial(gr.Image,
+        visible=False,
+    )(**kwargs)
+
+
+with gr.Blocks() as demo:
+    gr.Markdown(
+        f"""
+        # 🍹 bartab: Fitness from pooled competition assays
+
+        *Using* `bartab` v{bartab.__version__} | [Documentation](https://github.com/scbirlab/bartab) | [Tutorial on analysis principles](https://huggingface.co/spaces/scbirlab/tutorial-seq-fitness)
+        
+        Infer the competitive fitness of barcoded strains from next-generation 
+        sequencing of pooled growth experiments.
+        
+        Upload your count table, sample sheet, and barcode sheet, then 
+        click **Calculate fitness**.
+
+        """
+    )
+    gr.Markdown(
+        """
+        ---
+
+        ## 1️⃣ Input tables
+
+        Three tables are required. You can upload CSV, TSV, or XLSX files, 
+        or try one of the **example datasets** below.
+
+        """
+    )
+    input_filenames = {
+        "count_table": gr.Textbox(interactive=False, visible=False),
+        "sample_sheet": gr.Textbox(interactive=False, visible=False),
+        "strain_sheet": gr.Textbox(interactive=False, visible=False),
+    }
+    app_root = os.path.dirname(__file__)
+    data_path = os.path.join(app_root, "data", "examples", "single-point")
+    control_strains = {
+        "reference": _invisible_dropdown(
+            label="Reference (WT) barcode name",
+            render=False,
+        ),
+        "spike": _invisible_dropdown(
+            label="Spike-in barcode name (if using)",
+            render=False,
+        ),
+    }
+    analysis_opts = {
+        "use_spike": gr.Radio(
+            label="Culture expansion uses:",
+            choices=["Spike", "Growth"],
+            value="Spike",
+            render=False,
+        ),
+        "growth_type": gr.Radio(
+            label="Growth type",
+            choices=["density", "generations"],
+            value="density",
+            visible=False,
+            render=False,
+        ),
+        "pseudocount": gr.Number(
+            label="Pseudocount",
+            value=1.,
+            render=False,
+        ),
+    }
+    plotting_opts = {
+        "highlight": _invisible_dropdown(
+            label="Strain(s) to highlight in plots",
+            render=False,
+            multiselect=True,
+        ),
+        "controls": gr.Textbox(
+            label="Prefix of control barcode names",
+            value="ctrl_",
+            render=False,
+        ),
+    }
+    mode_switch = gr.Radio(
+        label="Analysis mode",
+        choices=list(MODES.values()),
+        value=MODES["single"],
+        render=False,
+    )
+    examples = gr.Examples(
+        label="Examples with synthetic data",
+        examples=[
+            [
+                os.path.join(data_path, "test_count.csv"),
+                os.path.join(data_path, "test_sample_meta.csv"),
+                os.path.join(data_path, "test_strain_meta.csv"),
+                MODES["single"],
+                "Spike",
+            ],
+            [
+                os.path.join(data_path, "test_count.csv"),
+                os.path.join(data_path, "test_sample_meta.csv"),
+                os.path.join(data_path, "test_strain_meta.csv"),
+                MODES["single"],
+                "Growth",
+            ],
+            [
+                os.path.join(data_path, "dose-response_count.csv"),
+                os.path.join(data_path, "dose-response_sample_meta.csv"),
+                os.path.join(data_path, "dose-response_strain_meta.csv"),
+                MODES["dose response"],
+                "Spike",
+            ],
+            [
+                os.path.join(data_path, "dose-response_count.csv"),
+                os.path.join(data_path, "dose-response_sample_meta.csv"),
+                os.path.join(data_path, "dose-response_strain_meta.csv"),
+                MODES["dose response"],
+                "Growth",
+            ],
+        ],
+        example_labels=[
+            ["Single point, using spike-in"],
+            ["Single point, using growth"],
+            ["Dose response, using spike-in"],
+            ["Dose response, using growth"],
+        ],
+        inputs=[
+            input_filenames["count_table"], 
+            input_filenames["sample_sheet"], 
+            input_filenames["strain_sheet"],
+            mode_switch,
+            analysis_opts["use_spike"],
+        ],
+        # cache_examples=True,
+        # cache_mode="eager",
+    )
+    input_files = {}
+    input_cols = {}
+    go_button = gr.Button(
+        value="🚀 Calculate fitness!",
+        interactive=False,
+        render=False,
+    )
+    with gr.Row():
+        with gr.Column():
+            gr.Markdown(
+                """
+                ---
+
+                ### 🧮 Count table
+
+                One row per barcode per sample. Must contain:
+                - a column of **barcode/strain identifiers** (matching your barcode sheet)
+                - a column of **sample identifiers** (matching your sample sheet)  
+                - a column of **read or UMI counts**
+
+                """
+            )
+            input_files["count_table"] = _file_input(
+                label="Upload your barcode sequencing counts data here",
+            )
+            input_cols["count_table"] = {
+                "count": (_invisible_dropdown(
+                    label="Counts column",
+                ), "numeric"),
+            }
+        with gr.Column():
+            gr.Markdown(
+                """
+                ---
+
+                ### 📶 Barcode information
+
+                One row per unique barcode. Must contain:
+                - a column of **barcode identifiers**
+
+                Optionally: any metadata about strains (gene targets, constructs, 
+                etc.). These will be carried through to the output.
+
+                """
+            )
+            input_files["strain_sheet"] = _file_input(
+                label="Upload your barcode information here",
+            )
+            input_cols["strain_sheet"] = {
+                "strain_id": (_invisible_dropdown(
+                    label="Barcode identifier column",
+                ), "string"),
+            }
+    with gr.Row():
+        gr.Markdown(
+            r"""
+            ---
+
+            ### 🧪 Sample sheet
+
+            One row per sample (sequencing library). Must contain:
+            - **Sample ID**: unique identifier matching the count table
+            - **Culture ID**: biological replicate identifier. Samples from 
+            the same culture share this label
+            - **Timepoint**: numeric timepoint values. The earliest timepoint 
+            is treated as $t_0$.
+
+            **For spike-in normalisation**: 
+            no extra columns needed. Just include your spike-in barcode in 
+            the count table and barcode sheet.
+
+            **For growth-based normalisation**: add a column of OD600, CFU/mL, 
+            or generation counts measured at each sample.
+
+            **For dose-response analysis**: add a column of inducer/drug 
+            concentrations. Samples with concentration = 0 are treated as 
+            uninduced controls.
+
+            **For adaptive-volume sampling** (if you took different volumes 
+            from each sample): add a column of sampled volumes.
+
+            """
+        )
+        input_files["sample_sheet"] = _file_input(
+            label="Upload your sample information here",
+        )
+    with gr.Row():
+        input_cols["sample_sheet"] = {
+            "timepoint": (_invisible_dropdown(
+                label="Timepoint column",
+            ), "any"),
+            "dose": (_invisible_dropdown(
+                label="Concentration column",
+            ), "numeric"),
+            "sample_id": (_invisible_dropdown(
+                label="Individual sample ID column",
+            ), "string"),
+            "replicate": (_invisible_dropdown(
+                label="Culture / biological replicate column",
+            ), "string"),
+            "volume": (_invisible_dropdown(
+                label="Volume column (if using)",
+            ), "numeric"),
+            "growth": (_invisible_dropdown(
+                label="Growth column (if using)",
+            ), "numeric"),            
+        }
+        
+    with gr.Row():
+        input_data = {
+            key: gr.Dataframe(
+                label=f"Input data: {key}",
+                max_height=50,
+                visible=False,
+                interactive=False,
+            ) for key in input_files
+        }
+
+    adata = gr.State()
+    with gr.Row():
+        with gr.Column():
+            gr.Markdown(
+                """
+                ---
+
+                ## 2️⃣ Control strains
+
+                - **Reference (WT)**: the strain relative to which all 
+                fitness values are calculated. Fitness = 1 by definition.
+                - **Spike-in**: a non-growing strain (e.g. heat-killed or 
+                plasmid-only) added at a fixed concentration before library 
+                preparation. Used to infer how much the reference strain has 
+                expanded between timepoints, removing the need for growth 
+                measurements. Leave blank if using growth measurements instead.
+
+                """
+            )
+            for key, val in control_strains.items():
+                val.render()
+        with gr.Column():
+            gr.Markdown(
+                """
+                ---
+
+                ## 3️⃣ Analysis options
+
+                - **Culture expansion**: choose **Spike** if you have a 
+                non-growing spike-in control, or **Growth** if you have 
+                OD600/CFU measurements.
+                - **Pseudocount**: added to all counts before log transformation 
+                to avoid log(0).
+                - **Analysis mode**: choose **Single concentration** for standard fitness 
+                screens, or **Dose response** if your sample sheet contains a concentration 
+                column. Dose response fitting uses a 2-parameter Hill model to estimate the 
+                IC₅₀ and maximum effectfor each barcode.
+
+                """
+            )
+            for key, val in analysis_opts.items():
+                val.render()
+        with gr.Column():
+            gr.Markdown(
+                """
+                ---
+                
+                ## Plotting options
+
+                """
+            )
+            for key, val in plotting_opts.items():
+                val.render()
+
+    with gr.Column():
+        mode_switch.render()
+        go_button.render()
+        
+    mode_switch.change(
+        lambda x: gr.update(value=x),
+        inputs=[mode_switch],
+        outputs=[go_button],
+    )
+    gr.Markdown(
+        r"""
+        ## 4️⃣ Results
+
+        Fitness values are estimated by weighted least squares regression of the log-ratio of each barcode against the reference strain, using the spike-in or growth measurements as the x-axis. 
+
+        **Key output columns**:
+        - For single concentration:
+            - `fitness`: relative fitness ($w_i / w_{wt}$). Values < 1 indicate growth disadvantage; > 1 indicates advantage.
+            - `fitness_low` / `fitness_high`: 95% confidence interval bounds.
+            - `slope_p`: p-value for the slope being different from 0 (i.e. fitness ≠ 1).
+        - For dose-response: 
+            - `log_ic50` (log₁₀ concentration at 50% inhibition) and `log_ic50_p`.
+
+        Results and the full annotated dataset (`.h5ad`) can be downloaded below.
+
+        """
+    )
+
+    plots = {}
+    with gr.Row():
+        plots |= {
+            "dose_response": (
+                _invisible_plot(label="Dose response"),
+                _plot_dose_response,
+            ),
+            "count_time": (
+                _invisible_plot(label="Time vs count"),
+                _plot_time_vs_count,
+            ),
+            "count_exp": (
+                _invisible_plot(label="Expansion vs count"),
+                _plot_expansion_vs_count,
+            ),
+        }
+    with gr.Row():
+        plots |= {
+            "count_exp": (
+                _invisible_plot(label="Time vs ratio"),
+                _plot_time_vs_ratio,
+            ),
+            "ratio_exp": (
+                _invisible_plot(label="Expansion vs ratio"),
+                _plot_expansion_vs_ratio,
+            ),
+        }
+    with gr.Row():
+        plots |= {
+            "pred_obs": (
+                _invisible_plot(label="Predicted vs observed"),
+                _plot_pred_vs_true,
+            ),
+            "volcano": (
+                _invisible_plot(label="Volcano"),
+                _plot_volcano,
+            ),
+        }
+    with gr.Row():
+        download = gr.DownloadButton(
+            label="Download parameters as CSV",
+            visible=False,
+        )
+        download_adata = gr.DownloadButton(
+            label="Download all analysis as .h5ad",
+            visible=False,
+        )
+    output_table = gr.Dataframe(
+        label="Fitted parameters",
+        # max_height=100,
+        visible=False,
+        interactive=False,
+    )
+
+    # ======
+    # EVENTS
+    # ======
+
+    for key, input_file in input_files.items():
+        input_columns = input_cols[key]
+        event_fn = {
+            "fn": partial(load_input_data, cols=input_columns),
+            "outputs": [input_data[key]] + [
+                col[0] if isinstance(col, tuple) else col 
+                for _, col in input_columns.items()
+            ],
+        }
+        input_filenames[key].change(
+            _load_from_file, 
+            inputs=[input_filenames[key]], 
+            outputs=[input_file],
+        ).then(
+            **event_fn, 
+            inputs=[input_filenames[key]], 
+        )
+        input_file.upload(
+            **event_fn, 
+            inputs=[input_file], 
+        )
+        
+    input_cols["strain_sheet"]["strain_id"][0].change(
+        load_barcode_names,
+        inputs=[
+            input_data["strain_sheet"], 
+            input_cols["strain_sheet"]["strain_id"][0],
+        ],
+        outputs=[
+            control_strains["reference"],
+            control_strains["spike"],
+            plotting_opts["highlight"],
+        ],
+    ).then(
+        lambda : gr.update(interactive=True),
+        inputs=[],
+        outputs=[go_button],
+    )
+
+    analysis_opts["use_spike"].change(
+        lambda x: gr.update(visible=x == "Growth"),
+        inputs=[analysis_opts["use_spike"]],
+        outputs=[analysis_opts["growth_type"]],
+    )
+
+    comptuation_inputs = (
+        [
+            f for _, f in input_data.items()
+        ] + [
+            opt[0] if isinstance(opt, tuple) else opt
+            for _, input_col in input_cols.items()
+            for _, opt in input_col.items()
+        ] + [
+            v for k, v in control_strains.items()
+        ] + [
+            v for k, v in analysis_opts.items()
+        ]
+    )
+
+    evt = go_button.click(
+        fn=do_analysis,
+        inputs=comptuation_inputs + [mode_switch],
+        outputs=[
+            output_table,
+            adata,
+        ],
+    ).then(
+        download_tables,
+        inputs=[output_table, adata],
+        outputs=[download, download_adata],
+    )
+    
+    for key, (p, fn) in  plots.items():
+        evt = evt.then(
+            fn,
+            inputs=[
+                adata, 
+                plotting_opts["highlight"], 
+                plotting_opts["controls"], 
+                mode_switch,
+            ],
+            outputs=[p],
+        )
+    
+if __name__ == "__main__":
+    demo.queue() 
+    demo.launch(share=True)

data/examples/single-point/dose-response_sample_meta.csv

@@ -0,0 +1,181 @@
+sample_id,replicate,timepoint,dose,growth,generations,is_t0
+dose_0-rep_0-t_0,dose_0-rep_0,0.0,0.12345679012345678,117055.0,0.0,True
+dose_0-rep_0-t_1,dose_0-rep_0,1.0,0.12345679012345678,234292.89384810615,1.0011266371943761,False
+dose_0-rep_0-t_2,dose_0-rep_0,2.0,0.12345679012345678,427061.5122706585,1.8672573245346546,False
+dose_0-rep_0-t_3,dose_0-rep_0,3.0,0.12345679012345678,668560.4452222767,2.5138714412935887,False
+dose_0-rep_0-t_4,dose_0-rep_0,4.0,0.12345679012345678,883476.5973974025,2.9160053589907093,False
+dose_0-rep_0-t_5,dose_0-rep_0,5.0,0.12345679012345678,1024302.3566620992,3.129383171737148,False
+dose_0-rep_1-t_0,dose_0-rep_1,0.0,0.12345679012345678,117055.0,0.0,True
+dose_0-rep_1-t_1,dose_0-rep_1,1.0,0.12345679012345678,234292.89384810615,1.0011266371943761,False
+dose_0-rep_1-t_2,dose_0-rep_1,2.0,0.12345679012345678,427061.5122706585,1.8672573245346546,False
+dose_0-rep_1-t_3,dose_0-rep_1,3.0,0.12345679012345678,668560.4452222767,2.5138714412935887,False
+dose_0-rep_1-t_4,dose_0-rep_1,4.0,0.12345679012345678,883476.5973974025,2.9160053589907093,False
+dose_0-rep_1-t_5,dose_0-rep_1,5.0,0.12345679012345678,1024302.3566620992,3.129383171737148,False
+dose_0-rep_2-t_0,dose_0-rep_2,0.0,0.12345679012345678,117055.0,0.0,True
+dose_0-rep_2-t_1,dose_0-rep_2,1.0,0.12345679012345678,234292.89384810615,1.0011266371943761,False
+dose_0-rep_2-t_2,dose_0-rep_2,2.0,0.12345679012345678,427061.5122706585,1.8672573245346546,False
+dose_0-rep_2-t_3,dose_0-rep_2,3.0,0.12345679012345678,668560.4452222767,2.5138714412935887,False
+dose_0-rep_2-t_4,dose_0-rep_2,4.0,0.12345679012345678,883476.5973974025,2.9160053589907093,False
+dose_0-rep_2-t_5,dose_0-rep_2,5.0,0.12345679012345678,1024302.3566620992,3.129383171737148,False
+dose_1-rep_0-t_0,dose_1-rep_0,0.0,0.20117262002701805,117055.0,0.0,True
+dose_1-rep_0-t_1,dose_1-rep_0,1.0,0.20117262002701805,222422.7611495618,0.9261178702440107,False
+dose_1-rep_0-t_2,dose_1-rep_0,2.0,0.20117262002701805,394457.30302182614,1.7526825875833112,False
+dose_1-rep_0-t_3,dose_1-rep_0,3.0,0.20117262002701805,618290.3220565522,2.401097863199693,False
+dose_1-rep_0-t_4,dose_1-rep_0,4.0,0.20117262002701805,833511.6554962726,2.832015812996294,False
+dose_1-rep_0-t_5,dose_1-rep_0,5.0,0.20117262002701805,988339.672526323,3.077820391955883,False
+dose_1-rep_1-t_0,dose_1-rep_1,0.0,0.20117262002701805,117055.0,0.0,True
+dose_1-rep_1-t_1,dose_1-rep_1,1.0,0.20117262002701805,222422.7611495618,0.9261178702440107,False
+dose_1-rep_1-t_2,dose_1-rep_1,2.0,0.20117262002701805,394457.30302182614,1.7526825875833112,False
+dose_1-rep_1-t_3,dose_1-rep_1,3.0,0.20117262002701805,618290.3220565522,2.401097863199693,False
+dose_1-rep_1-t_4,dose_1-rep_1,4.0,0.20117262002701805,833511.6554962726,2.832015812996294,False
+dose_1-rep_1-t_5,dose_1-rep_1,5.0,0.20117262002701805,988339.672526323,3.077820391955883,False
+dose_1-rep_2-t_0,dose_1-rep_2,0.0,0.20117262002701805,117055.0,0.0,True
+dose_1-rep_2-t_1,dose_1-rep_2,1.0,0.20117262002701805,222422.7611495618,0.9261178702440107,False
+dose_1-rep_2-t_2,dose_1-rep_2,2.0,0.20117262002701805,394457.30302182614,1.7526825875833112,False
+dose_1-rep_2-t_3,dose_1-rep_2,3.0,0.20117262002701805,618290.3220565522,2.401097863199693,False
+dose_1-rep_2-t_4,dose_1-rep_2,4.0,0.20117262002701805,833511.6554962726,2.832015812996294,False
+dose_1-rep_2-t_5,dose_1-rep_2,5.0,0.20117262002701805,988339.672526323,3.077820391955883,False
+dose_2-rep_0-t_0,dose_2-rep_0,0.0,0.3278104266931334,117055.0,0.0,True
+dose_2-rep_0-t_1,dose_2-rep_0,1.0,0.3278104266931334,208766.2945323125,0.834702246130147,False
+dose_2-rep_0-t_2,dose_2-rep_0,2.0,0.3278104266931334,356837.17092046584,1.6080793449317268,False
+dose_2-rep_0-t_3,dose_2-rep_0,3.0,0.3278104266931334,557154.5259169642,2.250890951794641,False
+dose_2-rep_0-t_4,dose_2-rep_0,4.0,0.3278104266931334,767351.9087930336,2.712701791269896,False
+dose_2-rep_0-t_5,dose_2-rep_0,5.0,0.3278104266931334,936068.5780837294,2.9994276677073946,False
+dose_2-rep_1-t_0,dose_2-rep_1,0.0,0.3278104266931334,117055.0,0.0,True
+dose_2-rep_1-t_1,dose_2-rep_1,1.0,0.3278104266931334,208766.2945323125,0.834702246130147,False
+dose_2-rep_1-t_2,dose_2-rep_1,2.0,0.3278104266931334,356837.17092046584,1.6080793449317268,False
+dose_2-rep_1-t_3,dose_2-rep_1,3.0,0.3278104266931334,557154.5259169642,2.250890951794641,False
+dose_2-rep_1-t_4,dose_2-rep_1,4.0,0.3278104266931334,767351.9087930336,2.712701791269896,False
+dose_2-rep_1-t_5,dose_2-rep_1,5.0,0.3278104266931334,936068.5780837294,2.9994276677073946,False
+dose_2-rep_2-t_0,dose_2-rep_2,0.0,0.3278104266931334,117055.0,0.0,True
+dose_2-rep_2-t_1,dose_2-rep_2,1.0,0.3278104266931334,208766.2945323125,0.834702246130147,False
+dose_2-rep_2-t_2,dose_2-rep_2,2.0,0.3278104266931334,356837.17092046584,1.6080793449317268,False
+dose_2-rep_2-t_3,dose_2-rep_2,3.0,0.3278104266931334,557154.5259169642,2.250890951794641,False
+dose_2-rep_2-t_4,dose_2-rep_2,4.0,0.3278104266931334,767351.9087930336,2.712701791269896,False
+dose_2-rep_2-t_5,dose_2-rep_2,5.0,0.3278104266931334,936068.5780837294,2.9994276677073946,False
+dose_3-rep_0-t_0,dose_3-rep_0,0.0,0.5341665075212624,117055.0,0.0,True
+dose_3-rep_0-t_1,dose_3-rep_0,1.0,0.5341665075212624,194167.85296808663,0.7301178030311807,False
+dose_3-rep_0-t_2,dose_3-rep_0,2.0,0.5341665075212624,316942.8764619754,1.437036282232122,False
+dose_3-rep_0-t_3,dose_3-rep_0,3.0,0.5341665075212624,489512.636121671,2.0641595411437166,False
+dose_3-rep_0-t_4,dose_3-rep_0,4.0,0.5341665075212624,687865.3222216408,2.5549395655675515,False
+dose_3-rep_0-t_5,dose_3-rep_0,5.0,0.5341665075212624,867383.0729686539,2.889482728766238,False
+dose_3-rep_1-t_0,dose_3-rep_1,0.0,0.5341665075212624,117055.0,0.0,True
+dose_3-rep_1-t_1,dose_3-rep_1,1.0,0.5341665075212624,194167.85296808663,0.7301178030311807,False
+dose_3-rep_1-t_2,dose_3-rep_1,2.0,0.5341665075212624,316942.8764619754,1.437036282232122,False
+dose_3-rep_1-t_3,dose_3-rep_1,3.0,0.5341665075212624,489512.636121671,2.0641595411437166,False
+dose_3-rep_1-t_4,dose_3-rep_1,4.0,0.5341665075212624,687865.3222216408,2.5549395655675515,False
+dose_3-rep_1-t_5,dose_3-rep_1,5.0,0.5341665075212624,867383.0729686539,2.889482728766238,False
+dose_3-rep_2-t_0,dose_3-rep_2,0.0,0.5341665075212624,117055.0,0.0,True
+dose_3-rep_2-t_1,dose_3-rep_2,1.0,0.5341665075212624,194167.85296808663,0.7301178030311807,False
+dose_3-rep_2-t_2,dose_3-rep_2,2.0,0.5341665075212624,316942.8764619754,1.437036282232122,False
+dose_3-rep_2-t_3,dose_3-rep_2,3.0,0.5341665075212624,489512.636121671,2.0641595411437166,False
+dose_3-rep_2-t_4,dose_3-rep_2,4.0,0.5341665075212624,687865.3222216408,2.5549395655675515,False
+dose_3-rep_2-t_5,dose_3-rep_2,5.0,0.5341665075212624,867383.0729686539,2.889482728766238,False
+dose_4-rep_0-t_0,dose_4-rep_0,0.0,0.870423374374747,117055.0,0.0,True
+dose_4-rep_0-t_1,dose_4-rep_0,1.0,0.870423374374747,179849.0377847039,0.619599880459945,False
+dose_4-rep_0-t_2,dose_4-rep_0,2.0,0.870423374374747,278757.0181696531,1.2518215672959792,False
+dose_4-rep_0-t_3,dose_4-rep_0,3.0,0.870423374374747,423504.06388361723,1.8551892530849372,False
+dose_4-rep_0-t_4,dose_4-rep_0,4.0,0.870423374374747,605732.9764283933,2.3714953941895147,False
+dose_4-rep_0-t_5,dose_4-rep_0,5.0,0.870423374374747,791779.1028853382,2.757911430955815,False
+dose_4-rep_1-t_0,dose_4-rep_1,0.0,0.870423374374747,117055.0,0.0,True
+dose_4-rep_1-t_1,dose_4-rep_1,1.0,0.870423374374747,179849.0377847039,0.619599880459945,False
+dose_4-rep_1-t_2,dose_4-rep_1,2.0,0.870423374374747,278757.0181696531,1.2518215672959792,False
+dose_4-rep_1-t_3,dose_4-rep_1,3.0,0.870423374374747,423504.06388361723,1.8551892530849372,False
+dose_4-rep_1-t_4,dose_4-rep_1,4.0,0.870423374374747,605732.9764283933,2.3714953941895147,False
+dose_4-rep_1-t_5,dose_4-rep_1,5.0,0.870423374374747,791779.1028853382,2.757911430955815,False
+dose_4-rep_2-t_0,dose_4-rep_2,0.0,0.870423374374747,117055.0,0.0,True
+dose_4-rep_2-t_1,dose_4-rep_2,1.0,0.870423374374747,179849.0377847039,0.619599880459945,False
+dose_4-rep_2-t_2,dose_4-rep_2,2.0,0.870423374374747,278757.0181696531,1.2518215672959792,False
+dose_4-rep_2-t_3,dose_4-rep_2,3.0,0.870423374374747,423504.06388361723,1.8551892530849372,False
+dose_4-rep_2-t_4,dose_4-rep_2,4.0,0.870423374374747,605732.9764283933,2.3714953941895147,False
+dose_4-rep_2-t_5,dose_4-rep_2,5.0,0.870423374374747,791779.1028853382,2.757911430955815,False
+dose_5-rep_0-t_0,dose_5-rep_0,0.0,1.41835334112138,117055.0,0.0,True
+dose_5-rep_0-t_1,dose_5-rep_0,1.0,1.41835334112138,167060.34358941315,0.5131827498868529,False
+dose_5-rep_0-t_2,dose_5-rep_0,2.0,1.41835334112138,246041.4256362762,1.0717146800456518,False
+dose_5-rep_0-t_3,dose_5-rep_0,3.0,1.41835334112138,367542.52779013116,1.6507246312376866,False
+dose_5-rep_0-t_4,dose_5-rep_0,4.0,1.41835334112138,535033.3697428543,2.1924423140004268,False
+dose_5-rep_0-t_5,dose_5-rep_0,5.0,1.41835334112138,726051.529930286,2.6328853835701116,False
+dose_5-rep_1-t_0,dose_5-rep_1,0.0,1.41835334112138,117055.0,0.0,True
+dose_5-rep_1-t_1,dose_5-rep_1,1.0,1.41835334112138,167060.34358941315,0.5131827498868529,False
+dose_5-rep_1-t_2,dose_5-rep_1,2.0,1.41835334112138,246041.4256362762,1.0717146800456518,False
+dose_5-rep_1-t_3,dose_5-rep_1,3.0,1.41835334112138,367542.52779013116,1.6507246312376866,False
+dose_5-rep_1-t_4,dose_5-rep_1,4.0,1.41835334112138,535033.3697428543,2.1924423140004268,False
+dose_5-rep_1-t_5,dose_5-rep_1,5.0,1.41835334112138,726051.529930286,2.6328853835701116,False
+dose_5-rep_2-t_0,dose_5-rep_2,0.0,1.41835334112138,117055.0,0.0,True
+dose_5-rep_2-t_1,dose_5-rep_2,1.0,1.41835334112138,167060.34358941315,0.5131827498868529,False
+dose_5-rep_2-t_2,dose_5-rep_2,2.0,1.41835334112138,246041.4256362762,1.0717146800456518,False
+dose_5-rep_2-t_3,dose_5-rep_2,3.0,1.41835334112138,367542.52779013116,1.6507246312376866,False
+dose_5-rep_2-t_4,dose_5-rep_2,4.0,1.41835334112138,535033.3697428543,2.1924423140004268,False
+dose_5-rep_2-t_5,dose_5-rep_2,5.0,1.41835334112138,726051.529930286,2.6328853835701116,False
+dose_6-rep_0-t_0,dose_6-rep_0,0.0,2.3112042478354495,117055.0,0.0,True
+dose_6-rep_0-t_1,dose_6-rep_0,1.0,2.3112042478354495,156640.12723312783,0.4202672814724111,False
+dose_6-rep_0-t_2,dose_6-rep_0,2.0,2.3112042478354495,220761.033587925,0.9152989848252058,False
+dose_6-rep_0-t_3,dose_6-rep_0,3.0,2.3112042478354495,325801.3065623647,1.4768058290896102,False
+dose_6-rep_0-t_4,dose_6-rep_0,4.0,2.3112042478354495,483650.1317704568,2.0467772315256103,False
+dose_6-rep_0-t_5,dose_6-rep_0,5.0,2.3112042478354495,680616.3580916729,2.5396552653559166,False
+dose_6-rep_1-t_0,dose_6-rep_1,0.0,2.3112042478354495,117055.0,0.0,True
+dose_6-rep_1-t_1,dose_6-rep_1,1.0,2.3112042478354495,156640.12723312783,0.4202672814724111,False
+dose_6-rep_1-t_2,dose_6-rep_1,2.0,2.3112042478354495,220761.033587925,0.9152989848252058,False
+dose_6-rep_1-t_3,dose_6-rep_1,3.0,2.3112042478354495,325801.3065623647,1.4768058290896102,False
+dose_6-rep_1-t_4,dose_6-rep_1,4.0,2.3112042478354495,483650.1317704568,2.0467772315256103,False
+dose_6-rep_1-t_5,dose_6-rep_1,5.0,2.3112042478354495,680616.3580916729,2.5396552653559166,False
+dose_6-rep_2-t_0,dose_6-rep_2,0.0,2.3112042478354495,117055.0,0.0,True
+dose_6-rep_2-t_1,dose_6-rep_2,1.0,2.3112042478354495,156640.12723312783,0.4202672814724111,False
+dose_6-rep_2-t_2,dose_6-rep_2,2.0,2.3112042478354495,220761.033587925,0.9152989848252058,False
+dose_6-rep_2-t_3,dose_6-rep_2,3.0,2.3112042478354495,325801.3065623647,1.4768058290896102,False
+dose_6-rep_2-t_4,dose_6-rep_2,4.0,2.3112042478354495,483650.1317704568,2.0467772315256103,False
+dose_6-rep_2-t_5,dose_6-rep_2,5.0,2.3112042478354495,680616.3580916729,2.5396552653559166,False
+dose_7-rep_0-t_0,dose_7-rep_0,0.0,3.7661032130325105,117055.0,0.0,True
+dose_7-rep_0-t_1,dose_7-rep_0,1.0,3.7661032130325105,148801.10472623174,0.3461986768823421,False
+dose_7-rep_0-t_2,dose_7-rep_0,2.0,3.7661032130325105,202753.24840351264,0.7925384682803503,False
+dose_7-rep_0-t_3,dose_7-rep_0,3.0,3.7661032130325105,297423.17062527494,1.3453304838789686,False
+dose_7-rep_0-t_4,dose_7-rep_0,4.0,3.7661032130325105,450412.9783225577,1.9440618377944692,False
+dose_7-rep_0-t_5,dose_7-rep_0,5.0,3.7661032130325105,653766.1815845432,2.4815881899955756,False
+dose_7-rep_1-t_0,dose_7-rep_1,0.0,3.7661032130325105,117055.0,0.0,True
+dose_7-rep_1-t_1,dose_7-rep_1,1.0,3.7661032130325105,148801.10472623174,0.3461986768823421,False
+dose_7-rep_1-t_2,dose_7-rep_1,2.0,3.7661032130325105,202753.24840351264,0.7925384682803503,False
+dose_7-rep_1-t_3,dose_7-rep_1,3.0,3.7661032130325105,297423.17062527494,1.3453304838789686,False
+dose_7-rep_1-t_4,dose_7-rep_1,4.0,3.7661032130325105,450412.9783225577,1.9440618377944692,False
+dose_7-rep_1-t_5,dose_7-rep_1,5.0,3.7661032130325105,653766.1815845432,2.4815881899955756,False
+dose_7-rep_2-t_0,dose_7-rep_2,0.0,3.7661032130325105,117055.0,0.0,True
+dose_7-rep_2-t_1,dose_7-rep_2,1.0,3.7661032130325105,148801.10472623174,0.3461986768823421,False
+dose_7-rep_2-t_2,dose_7-rep_2,2.0,3.7661032130325105,202753.24840351264,0.7925384682803503,False
+dose_7-rep_2-t_3,dose_7-rep_2,3.0,3.7661032130325105,297423.17062527494,1.3453304838789686,False
+dose_7-rep_2-t_4,dose_7-rep_2,4.0,3.7661032130325105,450412.9783225577,1.9440618377944692,False
+dose_7-rep_2-t_5,dose_7-rep_2,5.0,3.7661032130325105,653766.1815845432,2.4815881899955756,False
+dose_8-rep_0-t_0,dose_8-rep_0,0.0,6.1368584903291605,117055.0,0.0,True
+dose_8-rep_0-t_1,dose_8-rep_0,1.0,6.1368584903291605,143258.0596617452,0.29142974648746456,False
+dose_8-rep_0-t_2,dose_8-rep_0,2.0,6.1368584903291605,190613.57755750668,0.7034643267976026,False
+dose_8-rep_0-t_3,dose_8-rep_0,3.0,6.1368584903291605,279137.85937086184,1.2537912494702186,False
+dose_8-rep_0-t_4,dose_8-rep_0,4.0,6.1368584903291605,430135.55843049457,1.8776048404415195,False
+dose_8-rep_0-t_5,dose_8-rep_0,5.0,6.1368584903291605,639010.8809450462,2.448653936862557,False
+dose_8-rep_1-t_0,dose_8-rep_1,0.0,6.1368584903291605,117055.0,0.0,True
+dose_8-rep_1-t_1,dose_8-rep_1,1.0,6.1368584903291605,143258.0596617452,0.29142974648746456,False
+dose_8-rep_1-t_2,dose_8-rep_1,2.0,6.1368584903291605,190613.57755750668,0.7034643267976026,False
+dose_8-rep_1-t_3,dose_8-rep_1,3.0,6.1368584903291605,279137.85937086184,1.2537912494702186,False
+dose_8-rep_1-t_4,dose_8-rep_1,4.0,6.1368584903291605,430135.55843049457,1.8776048404415195,False
+dose_8-rep_1-t_5,dose_8-rep_1,5.0,6.1368584903291605,639010.8809450462,2.448653936862557,False
+dose_8-rep_2-t_0,dose_8-rep_2,0.0,6.1368584903291605,117055.0,0.0,True
+dose_8-rep_2-t_1,dose_8-rep_2,1.0,6.1368584903291605,143258.0596617452,0.29142974648746456,False
+dose_8-rep_2-t_2,dose_8-rep_2,2.0,6.1368584903291605,190613.57755750668,0.7034643267976026,False
+dose_8-rep_2-t_3,dose_8-rep_2,3.0,6.1368584903291605,279137.85937086184,1.2537912494702186,False
+dose_8-rep_2-t_4,dose_8-rep_2,4.0,6.1368584903291605,430135.55843049457,1.8776048404415195,False
+dose_8-rep_2-t_5,dose_8-rep_2,5.0,6.1368584903291605,639010.8809450462,2.448653936862557,False
+dose_9-rep_0-t_0,dose_9-rep_0,0.0,10.0,117055.0,0.0,True
+dose_9-rep_0-t_1,dose_9-rep_0,1.0,10.0,139507.74375037773,0.2531586444918018,False
+dose_9-rep_0-t_2,dose_9-rep_0,2.0,10.0,182703.5738878364,0.642318294551051,False
+dose_9-rep_0-t_3,dose_9-rep_0,3.0,10.0,267670.4959774961,1.1932715660229316,False
+dose_9-rep_0-t_4,dose_9-rep_0,4.0,10.0,418045.1524475581,1.8364722136788576,False
+dose_9-rep_0-t_5,dose_9-rep_0,5.0,10.0,631095.7950410473,2.4306724503882338,False
+dose_9-rep_1-t_0,dose_9-rep_1,0.0,10.0,117055.0,0.0,True
+dose_9-rep_1-t_1,dose_9-rep_1,1.0,10.0,139507.74375037773,0.2531586444918018,False
+dose_9-rep_1-t_2,dose_9-rep_1,2.0,10.0,182703.5738878364,0.642318294551051,False
+dose_9-rep_1-t_3,dose_9-rep_1,3.0,10.0,267670.4959774961,1.1932715660229316,False
+dose_9-rep_1-t_4,dose_9-rep_1,4.0,10.0,418045.1524475581,1.8364722136788576,False
+dose_9-rep_1-t_5,dose_9-rep_1,5.0,10.0,631095.7950410473,2.4306724503882338,False
+dose_9-rep_2-t_0,dose_9-rep_2,0.0,10.0,117055.0,0.0,True
+dose_9-rep_2-t_1,dose_9-rep_2,1.0,10.0,139507.74375037773,0.2531586444918018,False
+dose_9-rep_2-t_2,dose_9-rep_2,2.0,10.0,182703.5738878364,0.642318294551051,False
+dose_9-rep_2-t_3,dose_9-rep_2,3.0,10.0,267670.4959774961,1.1932715660229316,False
+dose_9-rep_2-t_4,dose_9-rep_2,4.0,10.0,418045.1524475581,1.8364722136788576,False
+dose_9-rep_2-t_5,dose_9-rep_2,5.0,10.0,631095.7950410473,2.4306724503882338,False

data/examples/single-point/dose-response_strain_meta.csv

@@ -0,0 +1,117 @@
+strain_id,dose_params,is_reference,is_spike
+wt,"[10000.0, 1.0]",True,False
+spike,"[1e-09, 0.0]",False,True
+mut_A,"[0.1, 0.0]",False,False
+mut_B,"[1.0, 0.0]",False,False
+mut_C,"[10.0, 0.0]",False,False
+mut_D,"[1.0, 0.5]",False,False
+ctrl_000,"[10000.0, 1.0]",False,False
+ctrl_001,"[10000.0, 1.0]",False,False
+ctrl_002,"[10000.0, 1.0]",False,False
+ctrl_003,"[10000.0, 1.0]",False,False
+ctrl_004,"[10000.0, 1.0]",False,False
+ctrl_005,"[10000.0, 1.0]",False,False
+ctrl_006,"[10000.0, 1.0]",False,False
+ctrl_007,"[10000.0, 1.0]",False,False
+ctrl_008,"[10000.0, 1.0]",False,False
+ctrl_009,"[10000.0, 1.0]",False,False
+str_000,"[np.float64(1.160934072833945), 0.0]",False,False
+str_001,"[np.float64(0.6583176596280784), 0.0]",False,False
+str_002,"[np.float64(1.2878968798670738), 0.0]",False,False
+str_003,"[np.float64(1.046052043589046), 0.0]",False,False
+str_004,"[np.float64(0.1412660218314743), 0.0]",False,False
+str_005,"[np.float64(1.463433527455134), 0.0]",False,False
+str_006,"[np.float64(1.1417095529855295), 0.0]",False,False
+str_007,"[np.float64(1.1790964579154308), 0.0]",False,False
+str_008,"[np.float64(0.1921704490133188), 0.0]",False,False
+str_009,"[np.float64(0.6755789068433506), 0.0]",False,False
+str_010,"[np.float64(0.5561970363488719), 0.0]",False,False
+str_011,"[np.float64(1.3901474832729028), 0.0]",False,False
+str_012,"[np.float64(0.9657976801209969), 0.0]",False,False
+str_013,"[np.float64(1.234142419906245), 0.0]",False,False
+str_014,"[np.float64(0.6651212982409966), 0.0]",False,False
+str_015,"[np.float64(0.34085808267716533), 0.0]",False,False
+str_016,"[np.float64(0.8318771805237521), 0.0]",False,False
+str_017,"[np.float64(0.09572588415626299), 0.0]",False,False
+str_018,"[np.float64(1.241446757988873), 0.0]",False,False
+str_019,"[np.float64(0.9474965986830972), 0.0]",False,False
+str_020,"[np.float64(1.1371316101280606), 0.0]",False,False
+str_021,"[np.float64(0.5317889521948025), 0.0]",False,False
+str_022,"[np.float64(1.4560470365923548), 0.0]",False,False
+str_023,"[np.float64(1.3396816819832966), 0.0]",False,False
+str_024,"[np.float64(1.1675752456106427), 0.0]",False,False
+str_025,"[np.float64(0.29195806177795136), 0.0]",False,False
+str_026,"[np.float64(0.7000815055905513), 0.0]",False,False
+str_027,"[np.float64(0.06570564868084317), 0.0]",False,False
+str_028,"[np.float64(0.23143423810132174), 0.0]",False,False
+str_029,"[np.float64(1.024573429863682), 0.0]",False,False
+str_030,"[np.float64(1.1171432338617258), 0.0]",False,False
+str_031,"[np.float64(1.451264598651315), 0.0]",False,False
+str_032,"[np.float64(0.4887380372072279), 0.0]",False,False
+str_033,"[np.float64(0.5556895590523033), 0.0]",False,False
+str_034,"[np.float64(0.7043337169137118), 0.0]",False,False
+str_035,"[np.float64(0.28420703862642854), 0.0]",False,False
+str_036,"[np.float64(0.19488225800320746), 0.0]",False,False
+str_037,"[np.float64(0.7135573893389007), 0.0]",False,False
+str_038,"[np.float64(0.34036402357632617), 0.0]",False,False
+str_039,"[np.float64(1.0047209920237656), 0.0]",False,False
+str_040,"[np.float64(0.6557278783084961), 0.0]",False,False
+str_041,"[np.float64(1.249017294086756), 0.0]",False,False
+str_042,"[np.float64(1.0503976530033736), 0.0]",False,False
+str_043,"[np.float64(0.4685499620730616), 0.0]",False,False
+str_044,"[np.float64(1.2483897020928016), 0.0]",False,False
+str_045,"[np.float64(1.2071465362452027), 0.0]",False,False
+str_046,"[np.float64(0.5812175685452616), 0.0]",False,False
+str_047,"[np.float64(0.43249215589536616), 0.0]",False,False
+str_048,"[np.float64(1.0237432559624633), 0.0]",False,False
+str_049,"[np.float64(0.2096287254139647), 0.0]",False,False
+str_050,"[np.float64(0.2998623037126625), 0.0]",False,False
+str_051,"[np.float64(0.011043404626508269), 0.0]",False,False
+str_052,"[np.float64(1.1803865662532076), 0.0]",False,False
+str_053,"[np.float64(0.9972762848880482), 0.0]",False,False
+str_054,"[np.float64(1.0577480679395026), 0.0]",False,False
+str_055,"[np.float64(1.1710935465329517), 0.0]",False,False
+str_056,"[np.float64(0.6883736633075099), 0.0]",False,False
+str_057,"[np.float64(0.8531117939293407), 0.0]",False,False
+str_058,"[np.float64(0.20969549719148617), 0.0]",False,False
+str_059,"[np.float64(0.17179511030396016), 0.0]",False,False
+str_060,"[np.float64(1.0026044426857075), 0.0]",False,False
+str_061,"[np.float64(0.7066443092146988), 0.0]",False,False
+str_062,"[np.float64(0.8478541597217832), 0.0]",False,False
+str_063,"[np.float64(1.1474982861240384), 0.0]",False,False
+str_064,"[np.float64(0.9520774800008862), 0.0]",False,False
+str_065,"[np.float64(0.8303691009869938), 0.0]",False,False
+str_066,"[np.float64(0.8388107411181203), 0.0]",False,False
+str_067,"[np.float64(0.4559251470939183), 0.0]",False,False
+str_068,"[np.float64(0.04622675185190911), 0.0]",False,False
+str_069,"[np.float64(0.6550760838485435), 0.0]",False,False
+str_070,"[np.float64(0.3218770092292938), 0.0]",False,False
+str_071,"[np.float64(0.6127929655869542), 0.0]",False,False
+str_072,"[np.float64(1.2801046099022493), 0.0]",False,False
+str_073,"[np.float64(0.3509092287980111), 0.0]",False,False
+str_074,"[np.float64(0.08745411253359903), 0.0]",False,False
+str_075,"[np.float64(0.4220758380329948), 0.0]",False,False
+str_076,"[np.float64(0.4403906366500254), 0.0]",False,False
+str_077,"[np.float64(0.9928747720903426), 0.0]",False,False
+str_078,"[np.float64(0.8355482285119176), 0.0]",False,False
+str_079,"[np.float64(1.1758473136596201), 0.0]",False,False
+str_080,"[np.float64(0.9964703104910813), 0.0]",False,False
+str_081,"[np.float64(0.6095802921601058), 0.0]",False,False
+str_082,"[np.float64(1.221030576999052), 0.0]",False,False
+str_083,"[np.float64(0.25045937986155586), 0.0]",False,False
+str_084,"[np.float64(0.034068109700790716), 0.0]",False,False
+str_085,"[np.float64(0.13507179116346263), 0.0]",False,False
+str_086,"[np.float64(1.0835390258946753), 0.0]",False,False
+str_087,"[np.float64(0.6928158453770811), 0.0]",False,False
+str_088,"[np.float64(0.2419076685504027), 0.0]",False,False
+str_089,"[np.float64(0.7515671626550453), 0.0]",False,False
+str_090,"[np.float64(0.22846815406975263), 0.0]",False,False
+str_091,"[np.float64(1.044480562616604), 0.0]",False,False
+str_092,"[np.float64(0.669234413361046), 0.0]",False,False
+str_093,"[np.float64(0.5715318391447237), 0.0]",False,False
+str_094,"[np.float64(0.4522681337218148), 0.0]",False,False
+str_095,"[np.float64(0.9454238896783327), 0.0]",False,False
+str_096,"[np.float64(0.5427189158300856), 0.0]",False,False
+str_097,"[np.float64(0.1314748789741515), 0.0]",False,False
+str_098,"[np.float64(0.17700885318077297), 0.0]",False,False
+str_099,"[np.float64(1.4428464968242718), 0.0]",False,False

data/examples/single-point/test_sample_meta.csv

@@ -0,0 +1,19 @@
+sample_id,replicate,timepoint,dose,growth,generations,is_t0
+dose_0-rep_0-t_0,dose_0-rep_0,0.0,1.0,117055.0,0.0,True
+dose_0-rep_0-t_1,dose_0-rep_0,1.0,1.0,235530.43107290982,1.008726910963631,False
+dose_0-rep_0-t_2,dose_0-rep_0,2.0,1.0,457470.31913837086,1.966491583356558,False
+dose_0-rep_0-t_3,dose_0-rep_0,3.0,1.0,745400.9751793981,2.6708301460588992,False
+dose_0-rep_0-t_4,dose_0-rep_0,4.0,1.0,972048.1257198368,3.053841182431035,False
+dose_0-rep_0-t_5,dose_0-rep_0,5.0,1.0,1088034.9223100767,3.216466397510134,False
+dose_0-rep_1-t_0,dose_0-rep_1,0.0,1.0,117055.0,0.0,True
+dose_0-rep_1-t_1,dose_0-rep_1,1.0,1.0,235530.43107290982,1.008726910963631,False
+dose_0-rep_1-t_2,dose_0-rep_1,2.0,1.0,457470.31913837086,1.966491583356558,False
+dose_0-rep_1-t_3,dose_0-rep_1,3.0,1.0,745400.9751793981,2.6708301460588992,False
+dose_0-rep_1-t_4,dose_0-rep_1,4.0,1.0,972048.1257198368,3.053841182431035,False
+dose_0-rep_1-t_5,dose_0-rep_1,5.0,1.0,1088034.9223100767,3.216466397510134,False
+dose_0-rep_2-t_0,dose_0-rep_2,0.0,1.0,117055.0,0.0,True
+dose_0-rep_2-t_1,dose_0-rep_2,1.0,1.0,235530.43107290982,1.008726910963631,False
+dose_0-rep_2-t_2,dose_0-rep_2,2.0,1.0,457470.31913837086,1.966491583356558,False
+dose_0-rep_2-t_3,dose_0-rep_2,3.0,1.0,745400.9751793981,2.6708301460588992,False
+dose_0-rep_2-t_4,dose_0-rep_2,4.0,1.0,972048.1257198368,3.053841182431035,False
+dose_0-rep_2-t_5,dose_0-rep_2,5.0,1.0,1088034.9223100767,3.216466397510134,False

data/examples/single-point/test_strain_meta.csv

@@ -0,0 +1,117 @@
+strain_id,true_fitness,is_reference,is_spike
+wt,1.0,True,False
+spike,0.0,False,True
+mut_A,1.5,False,False
+mut_B,0.5,False,False
+mut_C,0.9,False,False
+mut_D,1.1,False,False
+ctrl_000,1.0,False,False
+ctrl_001,1.0,False,False
+ctrl_002,1.0,False,False
+ctrl_003,1.0,False,False
+ctrl_004,1.0,False,False
+ctrl_005,1.0,False,False
+ctrl_006,1.0,False,False
+ctrl_007,1.0,False,False
+ctrl_008,1.0,False,False
+ctrl_009,1.0,False,False
+str_000,1.160934072833945,False,False
+str_001,0.6583176596280784,False,False
+str_002,1.2878968798670738,False,False
+str_003,1.046052043589046,False,False
+str_004,0.1412660218314743,False,False
+str_005,1.463433527455134,False,False
+str_006,1.1417095529855295,False,False
+str_007,1.1790964579154308,False,False
+str_008,0.1921704490133188,False,False
+str_009,0.6755789068433506,False,False
+str_010,0.5561970363488719,False,False
+str_011,1.3901474832729028,False,False
+str_012,0.9657976801209969,False,False
+str_013,1.234142419906245,False,False
+str_014,0.6651212982409966,False,False
+str_015,0.34085808267716533,False,False
+str_016,0.8318771805237521,False,False
+str_017,0.09572588415626299,False,False
+str_018,1.241446757988873,False,False
+str_019,0.9474965986830972,False,False
+str_020,1.1371316101280606,False,False
+str_021,0.5317889521948025,False,False
+str_022,1.4560470365923548,False,False
+str_023,1.3396816819832966,False,False
+str_024,1.1675752456106427,False,False
+str_025,0.29195806177795136,False,False
+str_026,0.7000815055905513,False,False
+str_027,0.06570564868084317,False,False
+str_028,0.23143423810132174,False,False
+str_029,1.024573429863682,False,False
+str_030,1.1171432338617258,False,False
+str_031,1.451264598651315,False,False
+str_032,0.4887380372072279,False,False
+str_033,0.5556895590523033,False,False
+str_034,0.7043337169137118,False,False
+str_035,0.28420703862642854,False,False
+str_036,0.19488225800320746,False,False
+str_037,0.7135573893389007,False,False
+str_038,0.34036402357632617,False,False
+str_039,1.0047209920237656,False,False
+str_040,0.6557278783084961,False,False
+str_041,1.249017294086756,False,False
+str_042,1.0503976530033736,False,False
+str_043,0.4685499620730616,False,False
+str_044,1.2483897020928016,False,False
+str_045,1.2071465362452027,False,False
+str_046,0.5812175685452616,False,False
+str_047,0.43249215589536616,False,False
+str_048,1.0237432559624633,False,False
+str_049,0.2096287254139647,False,False
+str_050,0.2998623037126625,False,False
+str_051,0.011043404626508269,False,False
+str_052,1.1803865662532076,False,False
+str_053,0.9972762848880482,False,False
+str_054,1.0577480679395026,False,False
+str_055,1.1710935465329517,False,False
+str_056,0.6883736633075099,False,False
+str_057,0.8531117939293407,False,False
+str_058,0.20969549719148617,False,False
+str_059,0.17179511030396016,False,False
+str_060,1.0026044426857075,False,False
+str_061,0.7066443092146988,False,False
+str_062,0.8478541597217832,False,False
+str_063,1.1474982861240384,False,False
+str_064,0.9520774800008862,False,False
+str_065,0.8303691009869938,False,False
+str_066,0.8388107411181203,False,False
+str_067,0.4559251470939183,False,False
+str_068,0.04622675185190911,False,False
+str_069,0.6550760838485435,False,False
+str_070,0.3218770092292938,False,False
+str_071,0.6127929655869542,False,False
+str_072,1.2801046099022493,False,False
+str_073,0.3509092287980111,False,False
+str_074,0.08745411253359903,False,False
+str_075,0.4220758380329948,False,False
+str_076,0.4403906366500254,False,False
+str_077,0.9928747720903426,False,False
+str_078,0.8355482285119176,False,False
+str_079,1.1758473136596201,False,False
+str_080,0.9964703104910813,False,False
+str_081,0.6095802921601058,False,False
+str_082,1.221030576999052,False,False
+str_083,0.25045937986155586,False,False
+str_084,0.034068109700790716,False,False
+str_085,0.13507179116346263,False,False
+str_086,1.0835390258946753,False,False
+str_087,0.6928158453770811,False,False
+str_088,0.2419076685504027,False,False
+str_089,0.7515671626550453,False,False
+str_090,0.22846815406975263,False,False
+str_091,1.044480562616604,False,False
+str_092,0.669234413361046,False,False
+str_093,0.5715318391447237,False,False
+str_094,0.4522681337218148,False,False
+str_095,0.9454238896783327,False,False
+str_096,0.5427189158300856,False,False
+str_097,0.1314748789741515,False,False
+str_098,0.17700885318077297,False,False
+str_099,1.4428464968242718,False,False

requirements.txt

@@ -0,0 +1,8 @@
+anndata
+carabiner-tools[pd,mpl]>=0.0.5.post3
+gradio==5.50.0
+nemony
+numpy
+scipy
+scikit-learn
+statsmodels