diff --git a/one/alf/path.py b/one/alf/path.py
index 0317d683..8b13bc0d 100644
--- a/one/alf/path.py
+++ b/one/alf/path.py
@@ -379,14 +379,12 @@ def get_alf_path(path: Union[str, pathlib.Path]) -> str:
     path = path.strip('/')
 
     # Check if session path
-    match_session = spec.regex(SESSION_SPEC).search(path)
-    if match_session:
+    if match_session := spec.regex(SESSION_SPEC).search(path):
         return path[match_session.start():]
 
     # Check if filename / relative path (i.e. collection + filename)
     parts = path.rsplit('/', 1)
-    match_filename = spec.regex(FILE_SPEC).match(parts[-1])
-    if match_filename:
+    if spec.regex(FILE_SPEC).match(parts[-1]):
         return path if spec.regex(f'{COLLECTION_SPEC}{FILE_SPEC}').match(path) else parts[-1]
 
 
diff --git a/one/api.py b/one/api.py
index 28a1c09a..c97196a2 100644
--- a/one/api.py
+++ b/one/api.py
@@ -14,6 +14,7 @@
 import time
 import threading
 import os
+import re
 
 import pandas as pd
 import numpy as np
@@ -2422,8 +2423,13 @@ def _download_aws(self, dsets, update_exists=True, keep_uuid=None, **_) -> List[
                     self._cache['_meta']['modified_time'] = datetime.now()
                 out_files.append(None)
                 continue
-            assert record['relative_path'].endswith(dset['rel_path']), \
-                f'Relative path for dataset {uuid} does not match Alyx record'
+            if 'relation' in dset:
+                # For non-session datasets the pandas record rel path is the full path
+                matches = dset['rel_path'].endswith(record['relative_path'])
+            else:
+                # For session datasets the pandas record rel path is relative to the session
+                matches = record['relative_path'].endswith(dset['rel_path'])
+            assert matches, f'Relative path for dataset {uuid} does not match Alyx record'
             source_path = PurePosixPath(record['data_repository_path'], record['relative_path'])
             local_path = self.cache_dir.joinpath(alfiles.get_alf_path(source_path))
             # Add UUIDs to filenames, if required
@@ -2532,6 +2538,9 @@ def _download_dataset(
         target_dir = []
         for x in valid_urls:
             _path = urllib.parse.urlsplit(x, allow_fragments=False).path.strip('/')
+            # Since rel_path for public FI file records starts with 'public/aggregates' instead of
+            # 'aggregates', we should discard the file path parts before 'aggregates' (if present)
+            _path = re.sub(r'^[\w\/]+(?=aggregates\/)', '', _path, count=1)
             target_dir.append(str(Path(cache_dir, alfiles.get_alf_path(_path)).parent))
         files = self._download_file(valid_urls, target_dir, **kwargs)
         # Return list of file paths or None if we failed to extract URL from dataset
diff --git a/one/converters.py b/one/converters.py
index 6879d0c3..3c040802 100644
--- a/one/converters.py
+++ b/one/converters.py
@@ -319,10 +319,15 @@ def record2url(self, record):
         else:
             raise TypeError(
                 f'record must be pandas.DataFrame or pandas.Series, got {type(record)} instead')
-        assert isinstance(record.name, tuple) and len(record.name) == 2
-        eid, uuid = record.name  # must be (eid, did)
-        session_path = self.eid2path(eid)
-        url = PurePosixALFPath(get_alf_path(session_path), record['rel_path'])
+        if 'session_path' in record:
+            # Check for session_path field (aggregate datasets have no eid in name)
+            session_path = record['session_path']
+            uuid = record.name if isinstance(record.name, str) else record.name[-1]
+        else:
+            assert isinstance(record.name, tuple) and len(record.name) == 2
+            eid, uuid = record.name  # must be (eid, did)
+            session_path = get_alf_path(self.eid2path(eid))
+        url = PurePosixALFPath(session_path, record['rel_path'])
         return webclient.rel_path2url(url.with_uuid(uuid).as_posix())
 
     def record2path(self, dataset) -> Optional[ALFPath]:
diff --git a/one/tests/test_one.py b/one/tests/test_one.py
index 661d8c01..c21862ed 100644
--- a/one/tests/test_one.py
+++ b/one/tests/test_one.py
@@ -29,7 +29,7 @@
 import datetime
 import logging
 import time
-from pathlib import Path
+from pathlib import Path, PurePosixPath
 from itertools import permutations, combinations_with_replacement
 from functools import partial
 import unittest
@@ -1423,6 +1423,30 @@ def test_load_aggregate(self):
         with self.assertRaises(alferr.ALFObjectNotFound):
             self.one.load_aggregate('subjects', 'ZM_1085', 'foo.bar')
 
+        # Test download file from HTTP dataserver
+        expected.unlink()
+        self.one.mode = 'remote'  # Can't download in local mode
+        with mock.patch.object(self.one, '_download_file', return_value=[expected]) as m, \
+                mock.patch.object(self.one, '_download_aws', side_effect=AssertionError):
+            file = self.one.load_aggregate('subjects', 'ZM_1085', dset, download_only=True)
+            # Check correct url passed to download_file
+            self.assertEqual(expected, file)
+            expected_src = (self.one.alyx._par.HTTP_DATA_SERVER +
+                            '/aggregates/Subjects/mainenlab/ZM_1085/' +
+                            '_ibl_subjectTraining.table.74dfb745-a7dc-4672-ace6-b556876c80cb.pqt')
+            expected_dst = str(file.parent)  # should be without 'public' part
+            m.assert_called_once_with([expected_src], [expected_dst], keep_uuid=False)
+        # Test download file from AWS
+        with mock.patch('one.remote.aws.s3_download_file', return_value=expected) as m, \
+                mock.patch.object(self.one, '_download_file', side_effect=AssertionError):
+            file = self.one.load_aggregate('subjects', 'ZM_1085', dset, download_only=True)
+            # Check correct url passed to download_file
+            self.assertEqual(expected, file)
+            expected_src = PurePosixPath(
+                expected_src[len(self.one.alyx._par.HTTP_DATA_SERVER) + 1:])
+            m.assert_called_once_with(
+                expected_src, expected, s3=mock.ANY, bucket_name='s3_bucket', overwrite=True)
+
     @classmethod
     def tearDownClass(cls) -> None:
         cls.tempdir.cleanup()