Fix for MSA block deletion

openfold/config.py

@@ -156,10 +156,12 @@ def model_config(
     elif name == "seqemb_initial_training":
         c.data.train.max_msa_clusters = 1
         c.data.eval.max_msa_clusters = 1
+        c.data.train.block_delete_msa = False
         c.data.train.max_distillation_msa_clusters = 1
     elif name == "seqemb_finetuning":
         c.data.train.max_msa_clusters = 1
         c.data.eval.max_msa_clusters = 1
+        c.data.train.block_delete_msa = False
         c.data.train.max_distillation_msa_clusters = 1
         c.data.train.crop_size = 384
         c.loss.violation.weight = 1.
@@ -311,6 +313,11 @@ def model_config(
                     "true_msa": [NUM_MSA_SEQ, NUM_RES],
                     "use_clamped_fape": [],
                 },
+                "block_delete_msa": {
+                    "msa_fraction_per_block": 0.3,
+                    "randomize_num_blocks": False,
+                    "num_blocks": 5,
+                },
                 "masked_msa": {
                     "profile_prob": 0.1,
                     "same_prob": 0.1,
@@ -355,6 +362,7 @@ def model_config(
             "predict": {
                 "fixed_size": True,
                 "subsample_templates": False,  # We want top templates.
+                "block_delete_msa": False,
                 "masked_msa_replace_fraction": 0.15,
                 "max_msa_clusters": 512,
                 "max_extra_msa": 1024,
@@ -368,6 +376,7 @@ def model_config(
             "eval": {
                 "fixed_size": True,
                 "subsample_templates": False,  # We want top templates.
+                "block_delete_msa": False,
                 "masked_msa_replace_fraction": 0.15,
                 "max_msa_clusters": 128,
                 "max_extra_msa": 1024,
@@ -381,6 +390,7 @@ def model_config(
             "train": {
                 "fixed_size": True,
                 "subsample_templates": True,
+                "block_delete_msa": True,
                 "masked_msa_replace_fraction": 0.15,
                 "max_msa_clusters": 128,
                 "max_extra_msa": 1024,

openfold/data/data_transforms.py

@@ -253,28 +253,33 @@ def block_delete_msa(protein, config):
         * config.msa_fraction_per_block
     ).to(torch.int32)
 
+    if int(block_num_seq) == 0:
+        return protein
+
     if config.randomize_num_blocks:
-        nb = torch.distributions.uniform.Uniform(
-            0, config.num_blocks + 1
-        ).sample()
+        nb = int(torch.randint(
+            low=0,
+            high=config.num_blocks + 1,
+            size=(1,),
+            device=protein["msa"].device,
+        )[0])
     else:
         nb = config.num_blocks
 
-    del_block_starts = torch.distributions.Uniform(0, num_seq).sample(nb)
-    del_blocks = del_block_starts[:, None] + torch.range(block_num_seq)
-    del_blocks = torch.clip(del_blocks, 0, num_seq - 1)
-    del_indices = torch.unique(torch.sort(torch.reshape(del_blocks, [-1])))[0]
+    del_block_starts = torch.randint(low=1, high=num_seq, size=(nb,), device=protein["msa"].device)
+    del_blocks = del_block_starts[:, None] + torch.arange(start=0, end=block_num_seq)
+    del_blocks = torch.clip(del_blocks, 1, num_seq - 1)
+    del_indices = torch.unique(torch.reshape(del_blocks, [-1]))
 
     # Make sure we keep the original sequence
-    combined = torch.cat((torch.range(1, num_seq)[None], del_indices[None]))
+    combined = torch.cat((torch.arange(start=0, end=num_seq), del_indices)).long()
     uniques, counts = combined.unique(return_counts=True)
-    difference = uniques[counts == 1]
-    intersection = uniques[counts > 1]
-    keep_indices = torch.squeeze(difference, 0)
+    keep_indices = uniques[counts == 1]
 
+    assert int(keep_indices[0]) == 0
     for k in MSA_FEATURE_NAMES:
         if k in protein:
-            protein[k] = torch.gather(protein[k], keep_indices)
+            protein[k] = torch.index_select(protein[k], 0, keep_indices)
 
     return protein
 

openfold/data/input_pipeline.py

@@ -71,6 +71,9 @@ def ensembled_transform_fns(common_cfg, mode_cfg, ensemble_seed):
     """Input pipeline data transformers that can be ensembled and averaged."""
     transforms = []
 
+    if mode_cfg.block_delete_msa:
+        transforms.append(data_transforms.block_delete_msa(common_cfg.block_delete_msa))
+
     if "max_distillation_msa_clusters" in mode_cfg:
         transforms.append(
             data_transforms.sample_msa_distillation(