diff --git a/mdevaluate/reader.py b/mdevaluate/reader.py
index 598072d..830ecb6 100755
--- a/mdevaluate/reader.py
+++ b/mdevaluate/reader.py
@@ -24,7 +24,6 @@ import numpy as np
 import MDAnalysis as mdanalysis
 from scipy import sparse
 from dask import delayed, __version__ as DASK_VERSION
-import pandas as pd
 import re
 
 
@@ -253,53 +252,12 @@ def correct_nojump_matrixes_for_whole(trajectory):
     save_nojump_matrixes(reader)
 
 
-def energy_reader(file, energies=None):
-    """Reads an gromacs energy file and output the data in a pandas DataFrame.
+def energy_reader(file):
+    """Reads a gromacs energy file with mdanalysis and returns an auxiliary file.
     Args:
         file: Filename of the energy file
-        energies (opt.): Specify energies to extract from the energy file
     """
-    if energies is None:
-        energies = np.arange(1, 100).astype("str")
-    directory = file.rsplit("/", 1)[0]
-    ps = subprocess.Popen(("echo", *energies), stdout=subprocess.PIPE)
-    try:
-        subprocess.run(
-            (
-                "gmx",
-                "energy",
-                "-f",
-                file,
-                "-o",
-                f"{directory}/tmp.xvg",
-                "-quiet",
-                "-nobackup",
-            ),
-            stdin=ps.stdout,
-        )
-    except FileNotFoundError:
-        print("No GROMACS found!")
-    ps.wait()
-    labels = []
-    is_legend = False
-    with open(f"{directory}/tmp.xvg") as f:
-        for i, line in enumerate(f):
-            if line.split(" ")[0] == "@":
-                if re.search("s\d+", line.split()[1]):
-                    is_legend = True
-                    labels.append(line.split('"')[1])
-            elif is_legend:
-                header = i
-                break
-
-    data = np.loadtxt(f"{directory}/tmp.xvg", skiprows=header)
-
-    df = pd.DataFrame({"Time": data[:, 0]})
-    for i, label in enumerate(labels):
-        tmp_df = pd.DataFrame({label: data[:, i + 1]})
-        df = pd.concat([df, tmp_df], axis=1)
-    subprocess.run(("rm", f"{directory}/tmp.xvg"))
-    return df
+    return mdanalysis.auxiliary.EDR.EDRReader(file)
 
 
 class BaseReader: