diff --git a/setup.py b/setup.py
index 4d69d4c..5b3f06b 100644
--- a/setup.py
+++ b/setup.py
@@ -4,7 +4,7 @@
inst_requires = ['numpy>=1.10',
'pandas>=0.20',
'seaborn>=0.7',
- 'scikit-learn>=0.18',
+ 'scikit-learn==0.19.1',
'pyfaidx>=0.5.4',
'pysam>=0.10.0',
'HTSeq>=0.6',
diff --git a/smallrnaseq/data/styles.css b/smallrnaseq/data/styles.css
index 8e0160d..79c7ab9 100644
--- a/smallrnaseq/data/styles.css
+++ b/smallrnaseq/data/styles.css
@@ -58,7 +58,7 @@ hr
margin-top: 30px;
padding: 1em 1.25em;
background-color: #fff;
- width: 70%;
+ width: 65%;
float: right;
overflow: hidden;
}
@@ -127,9 +127,9 @@ a:active { color: red; }
padding: 5px;
margin-top: 40px;
margin-right: 5px;
- float: right;
+ float: left;
height: 650px;
- width: 25%;
+ width: 30%;
position: fixed;
overflow:hidden;
top: 30;
@@ -147,8 +147,39 @@ a:active { color: red; }
.sidebar td, th {
text-align: left;
- font-size: 12px;
+ font-size: 11px;
+}
+
+/* Sortable tables */
+.sortable {
+ margin:0px;padding:0px;
+ width:100%;
+ border:1px solid #b2a7a7;
+ font-family: monospace;
+ font-size:11px;
+}
+.sortable table{
+ table-layout: fixed;
+ border-collapse: collapse;
+ border-spacing: 0;
+ height:100%;
+ margin:0px;padding:2px;
}
+.sortable td{
+ font-size: 11px;
+ border-width: 0px 0px 0px 0px;
+ margin: 1px;
+ padding: 2px 3px;
+ width: 100%;
+}
+.sortable th{
+ background-color: #ECF0F1;
+ border-width: 0px 0px 0px 0px;
+ max-width: 300px;
+ min-width: 40px;
+ word-wrap: break-word;
+}
+.sortable tr:nth-child(even){ background-color:#F7F7F7; }
.main, .aside
{
diff --git a/smallrnaseq/novel.py b/smallrnaseq/novel.py
index 3bc68f5..4939c75 100644
--- a/smallrnaseq/novel.py
+++ b/smallrnaseq/novel.py
@@ -362,6 +362,9 @@ def build_classifier(known, neg):
def precursor_classifier():
"""Get the stored miRNA precursor classifier model"""
+ #avoid being flooded with userwarnings
+ import warnings
+ warnings.filterwarnings("ignore", category=UserWarning)
from sklearn.externals import joblib
rf = joblib.load(os.path.join(datadir, 'premirna_model.joblib'))
return rf
@@ -432,7 +435,10 @@ def get_read_clusters(reads, cluster_distance=0, min_size=3, key='align_id'):
c = c.groupby(['strand']).filter(lambda x: len(x) > 1)
groups.append(c)
i+=1
- df = pd.concat(groups)
+ if len(groups) > 0:
+ df = pd.concat(groups)
+ else:
+ df = pd.DataFrame()
return df
def get_cluster_groups(rcl):
@@ -707,6 +713,9 @@ def find_mirnas(reads, ref_fasta, score_cutoff=.8, read_cutoff=50, species='',
rcl=rcl, ref_fasta=ref_fasta, score_cutoff=score_cutoff,
read_cutoff=read_cutoff)
+ if len(new) == 0:
+ print ('no precursors found above cutoff')
+ return None, None
new['seed'] = new.apply(lambda x: x.mature[1:7], 1)
#get coords column
new['coords'] = new.apply(get_coords_string,1)
@@ -720,6 +729,7 @@ def find_mirnas(reads, ref_fasta, score_cutoff=.8, read_cutoff=50, species='',
u = summarize(new)
print ('found %s unique novel mirnas' %len(u))
+ print ('score cutoff=%s' %score_cutoff)
print ('%s with known mature sequences' %len(u[-u.known_id.isnull()]))
#also return all the reads found in clusters
#found = pd.concat(X)
@@ -836,15 +846,16 @@ def create_report(df, reads, species=None, outfile='report.html'):
css = get_css()
h = ' novel miRNA'
h += '' %css
+ h += ''
h += ''
h += ''
h += ''
h += ''