Push to 0.0.10:

- improve app UI
- update packages

Dockerfile

@@ -21,3 +21,5 @@ RUN pip install -U /tmp/scoary/scoary_2-$SCOARY_VERSION-py3-none-any.whl && \
 ENV NUMBA_CACHE_DIR=/tmp/NUMBA_CACHE_DIR
 ENV CONFINT_DB=/tmp/CONFINT_DB
 ENV MPLCONFIGDIR=/tmp/MPLCONFIGDIR
+
+WORKDIR /data

README.md

@@ -17,19 +17,6 @@ association per trait. The results can be explored interactively with a simple,
 - [Understanding the p values](https://github.com/MrTomRod/scoary-2/wiki/Understanding-the-p-values)
 - [Usage as Python library](https://github.com/MrTomRod/scoary-2/wiki/Usage-as-Python-library)
 
-# Todo:
-
-- [X] Binarize traits during multiprocessing
-- [X] Idea: min_qval * min_empirical_p as new score
-- [X] Improve messaging and logging
-- [X] Remove multiple_testing_picking, more like Scoary 1 syntax?
-- [X] Log runtime?
-- [X] Refactoring
-- [X] `overview.html`: go to function, popover always left
-- [X] `overview.html` & `trait.html`: ~~set proper title~~, add logo/favicon
-- [ ] `overview.html` & `trait.html`: link to `summary.tsv` / `coverage-matrix.tsv` / `result.tsv` download
-- [X] Create wiki
-
 
 [orthofinder]: https://github.com/davidemms/OrthoFinder/
 [roary]: https://sanger-pathogens.github.io/Roary/

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "scoary-2"
-version = "0.0.9"
+version = "0.0.10"
 description = "Scoary 2: Associate genes with traits!"
 authors = ["MrTomRod <[email protected]>"]
 license = "GPL3"

scoary/load_traits.py

@@ -112,7 +112,7 @@ def apply_kmeans(kmeans: KMeans, data: pd.Series) -> (pd.Series, dict):
     # insert_positions = [v - i for i, v in enumerate(np.where(nan_vector)[0])]
     label_to_class = dict(zip(nan_free_data.index, labels_))
     return pd.array([label_to_class.get(l, pd.NA) for l in data.index], dtype="boolean"), \
-           {'method': 'kmeans', 'cutoff': np.mean(fit.cluster_centers_)}
+        {'method': 'kmeans', 'cutoff': np.mean(fit.cluster_centers_)}
 
 
 def classify_gm(val: [float, float], certainty_cutoff: float):
@@ -290,7 +290,7 @@ def worker(
     new_ns = grasp_namespace(BinarizeTraitNamespace, ns)
     del ns
 
-    kmeans = KMeans(n_clusters=2, random_state=new_ns.random_state)
+    kmeans = KMeans(n_init='auto', n_clusters=2, random_state=new_ns.random_state)
     gm = GaussianMixture(n_components=2, random_state=new_ns.random_state, covariance_type=new_ns.covariance_type)
 
     if new_ns.method == 'kmeans':

scoary/scoary.py

@@ -38,10 +38,10 @@ def scoary(
     :param traits: Path to trait presence/absence table: columns=traits, rows=isolates
     :param outdir: Directory to place output files
     :param multiple_testing: "method:cutoff" for filtering genes after Fisher's test, where cutoff is a number that
-    specifies the FWER and method is one of [bonferroni, sidak, holm-sidak, holm, simes-hochberg, hommel, fdr_bh,
-    fdr_by,  fdr_tsbh, fdr_tsbky]. Alternatively, the method can be 'native': then, the cutoff targets the uncorrected
-    pvalue from Fisher's test.
-    :param worst_cutoff: Drop traits if no gene with "worst" pvalue lower than threshold. Recommended if
+    specifies the FWER and method is one of [native, bonferroni, sidak, holm-sidak, holm, simes-hochberg, hommel,
+    fdr_bh, fdr_by,  fdr_tsbh, fdr_tsbky]. Alternatively, the method can be 'native': then, the cutoff targets the
+    uncorrected p-value from Fisher's test.
+    :param worst_cutoff: Drop traits if no gene with "worst" p-value lower than threshold. Recommended if
     dataset contains multiple species
     :param max_genes: Keep only n highest-scoring genes in Fisher's test. Recommended if dataset is big and contains
      multiple species; avoids waisting computational resources on traits that simply correlate with phylogeny

scoary/templates/overview.css

@@ -1,71 +1,91 @@
-/*#popoverTable, th , td{*/
-/*    width: 100%;*/
-/*    border: 1px solid;*/
-/*}*/
+#grid {
+    display: grid;
+    width: 100%;
+    height: 100vh;
+    grid-template-columns: 4fr 5fr;
+}
 
 .left {
-    /*float: left;*/
+    height: 100vh;
+    min-width: 610px;
+    max-width: 770px;
+    overflow: scroll;
 }
 
 .right {
-    float: right
+    height: 100vh;
+    min-width: 400px;
+    display: flex;
+    flex-direction: column;
+    justify-content: space-between;
+    overflow-y: auto;
 }
 
-.right-container {
-    position: fixed;
-    display: grid;
-    right: 0;
-    transform: translateY(-50%);
-    top: 50%;
+.btn {
+    margin-top: 5px;
+}
+
+#trait-select {
+    padding: 1rem;
+    margin-bottom: 1rem;
 }
 
-.popover {
-    max-width: 550px;
-    max-height: 100%;
-    overflow: auto;
+#trait-info {
+    margin-bottom: 1rem;
+    overflow-y: scroll;
 }
 
-.popover-header a {
-    color: unset
+#trait-info h2 a {
+    color: unset;
+    text-decoration: unset;
+    display: flex;
+    justify-content: space-between;
+    margin: 0;
+    padding: 0;
 }
 
-#highlighter {
-    width: 12rem;
+#trait-info h2 button {
+    margin: 0;
+}
+
+#svg-container {
+    overflow-y: scroll;
+    overflow-x: hidden;
 }
 
 .placeholder {
     opacity: unset;
     background-color: unset;
 }
 
-/* imitate bootstrap container shrinking */
-@media (min-width: 576px) {
-    .right-container {
-        max-width: 540px;
-    }
+#legend {
+    margin-bottom: 0.7rem;
 }
 
-@media (min-width: 768px) {
-    .right-container {
-        max-width: 720px;
-    }
+#legend td, #legend th {
+    text-align: center;
+    padding: 0 3px 0 3px;
 }
 
-@media (min-width: 992px) {
-    .right-container {
-        max-width: 960px;
-    }
+#legend th {
+    font-weight: bold;
 }
 
+#legend-link {
+    color: unset;
+    text-decoration: unset;
+}
 
-@media (min-width: 1200px) {
-    .right-container {
-        max-width: 1140px;
-    }
+#legend-link h3 {
+    text-align: center;
 }
 
-@media (min-width: 1400px) {
-    .right-container {
-        max-width: 1320px;
-    }
+.legend-title {
+    display: flex;
 }
+
+.legend-title .btn {
+    padding: 1px 8px 1px 8px;
+    margin-left: 5px;
+    margin-top: 2px;
+}

scoary/templates/overview.html

@@ -8,16 +8,15 @@
     <script src="https://cdnjs.cloudflare.com/ajax/libs/PapaParse/5.3.2/papaparse.min.js"
             crossorigin="anonymous"></script>
 
-    <link href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" rel="stylesheet"
-          integrity="sha384-1BmE4kWBq78iYhFldvKuhfTAU6auU8tT94WrHftjDbrCEXSU1oBoqyl2QvZ6jIW3" crossorigin="anonymous">
-
+    <link href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" rel="stylesheet"
+          integrity="sha384-rbsA2VBKQhggwzxH7pPCaAqO46MgnOM80zW1RWuH61DGLwZJEdK2Kadq2F9CUG65" crossorigin="anonymous">
     <link href="app/overview.css" rel="stylesheet">
 
     <script src="https://cdn.jsdelivr.net/npm/@popperjs/[email protected]/dist/umd/popper.min.js"
             crossorigin="anonymous"></script>
 
-    <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.1.3/dist/js/bootstrap.bundle.min.js"
-            integrity="sha384-ka7Sk0Gln4gmtz2MlQnikT1wXgYsOg+OMhuP+IlRH9sENBO0LRn5q+8nbTov4+1p"
+    <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.2.3/dist/js/bootstrap.bundle.min.js"
+            integrity="sha384-kenU1KFdBIe4zVF0s0G1M5b4hcpxyD9F7jL+jjXkk+Q2h455rYXK/7HAuoJl+0I4"
             crossorigin="anonymous"></script>
 
     <script src="https://cdnjs.cloudflare.com/ajax/libs/slim-select/1.27.1/slimselect.min.js"></script>
@@ -27,21 +26,53 @@
 
 <body>
 
-<section class="container">
-
-    <h1>Traits dendrogram</h1>
-
-    <div class="right-container">
-        <object type="image/svg+xml" data="overview_colorbar.svg" id="traits-colorbar" class="right"></object>
-
-        <select id="slim-select"></select>
-        <button id="copy-button" type="button" class="btn btn-info">Copy selected traits</button>
-        <button id="deselect-button" type="button" class="btn btn btn-warning">Deselect all</button>
-    </div>
-
-
-    <div id="svg-container">
-        <!-- <object type="image/svg+xml" data="overview_plot.svg" id="traits-dendrogram" class="left"></object> -->
+<section style="height: 100vh">
+    <div id="grid">
+        <div class="left">
+            <h1>Traits dendrogram</h1>
+            <div id="svg-container" class="col"></div>
+        </div>
+        <div class="right container">
+            <div>
+                <div id="trait-select">
+                    <select id="slim-select"></select>
+                    <div class="d-flex justify-content-evenly">
+                        <button id="copy-button" type="button" class="btn btn-primary">Copy selected traits</button>
+                        <button id="goto-button" type="button" class="btn btn-success">Scroll to trait</button>
+                        <button id="deselect-button" type="button" class="btn btn btn-warning">Deselect all</button>
+                    </div>
+                </div>
+
+                <div id="trait-info"></div>
+
+            </div>
+
+            <div class="d-flex flex-column align-items-center" title="Click here to learn more about these values!">
+
+                <div class="legend-title">
+                    <h3>Legend</h3>
+                    <a id="legend-link" href="https://github.com/MrTomRod/scoary-2/wiki/Understanding-the-p-values">
+                        <button type="button" class="btn btn-info">?</button>
+                    </a>
+                </div>
+
+                <p>The plot is log<sub>10</sub>-scaled.</p>
+
+                <table id="legend">
+                    <tr>
+                        <th style="color:#2ca02c">f</th>
+                        <th style="color:#9467bd">e</th>
+                        <th style="color:#d62728">✱</th>
+                    </tr>
+                    <tr>
+                        <td>Fisher's <i>q</i>-value</td>
+                        <td>Empirical <i>p</i>-value</td>
+                        <td><i>fq*ep</i> score</td>
+                    </tr>
+                </table>
+
+            </div>
+        </div>
     </div>
 
 </section>