diff --git a/.gitattributes b/.gitattributes
index caf44b62bb64c484d9aa0ce1ea318f55ed22d659..815d647c950d50c0c8e277f161f96dbd0244a898 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -300,3 +300,28 @@ data/clustering/model=NeuML__pubmedbert-base-embeddings/min=5/t=0.6_ALL_FIELDS.c
 data/clustering/model=NeuML__pubmedbert-base-embeddings/min=5/t=0.7_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
 data/clustering/model=NeuML__pubmedbert-base-embeddings/min=5/t=0.8_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
 data/clustering/model=NeuML__pubmedbert-base-embeddings/min=5/t=0.9_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
+data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_ALL_FIELDS.csv.gz filter=lfs diff=lfs merge=lfs -text
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_ALL_FIELDS.csv.gz
index 10e1cbcf79128508f40fcad923db03fe02cf63e2..37d2e6e03557dd328769b121ec8dc802728403e1 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8083354c1d344969a78e05e9b48b79edd838c773bc496f218ffef29819798769
-size 1628830
+oid sha256:fe5a8d6b523ddd9fe584e30558e454ff77140a4739080258e87e03826b601321
+size 2530445
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_FIELD_CLUSTER_METRICS.csv
index 95b1b55bc22c27e8a54aeac1bf9a478e26b27327..5c1b2448389be1cc43b40cff4dc6b3f8b4eecfeb 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.5_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6180,36,0.6571511626243591,1.651187615216294,10,0.5
-strain,2444,2362,37,0.5469115972518921,2.7834285481973353,10,0.5
-induction_temperature,3656,3652,6,0.4439533054828644,2.2633224841559603,10,0.5
-expression_strain,5677,5639,33,0.36717814207077026,2.430813075061987,10,0.5
-inducer,7889,7839,12,0.271083801984787,2.837082420210859,10,0.5
-desalting_process,3720,3703,7,0.21009418368339539,2.784306913963109,10,0.5
-lysis_buffer,3714,3705,9,0.199936181306839,2.6244682814745257,10,0.5
-elution_buffer,3530,3518,15,0.07715681940317154,2.8877032223343497,10,0.5
-plasmid,5480,5449,35,0.0373564176261425,3.900018759595488,10,0.5
-medium_name,4806,4771,42,-0.0426754392683506,3.121550043910463,10,0.5
-enzyme_name,7975,7826,155,-0.059842441231012344,3.933137631668066,10,0.5
-molecular_weight,295,294,1,,,10,0.5
-source_key,8021,8021,1,,,10,0.5
+organism_source,9834,9708,41,0.7405863404273987,1.7247533699740398,10,0.5
+induction_temperature,5383,5365,7,0.43037179112434387,2.2171575186292505,10,0.5
+expression_strain,8268,8230,35,0.31337201595306396,2.5496154253573597,10,0.5
+inducer,11954,11907,18,0.2553367614746094,3.1753845431392342,10,0.5
+lysis_buffer,5625,5614,12,0.2010677009820938,2.7359158138749766,10,0.5
+desalting_process,5540,5526,9,0.19206033647060394,3.3501053619583643,10,0.5
+elution_buffer,5198,5171,17,0.08386193215847015,3.06389209505366,10,0.5
+strain,3834,3739,40,0.0454237125813961,3.046488468393107,10,0.5
+plasmid,8217,8166,44,0.037487756460905075,4.467212089079907,10,0.5
+medium_name,7644,7598,53,-0.05805343762040138,3.3404684755222656,10,0.5
+enzyme_name,12099,11937,209,-0.06390286982059479,3.954636362204377,10,0.5
+molecular_weight,495,491,1,,,10,0.5
+source_key,12179,12179,1,,,10,0.5
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_ALL_FIELDS.csv.gz
index 8dd4bcc1624daf102bcbdb2bc8ec854ac4ee6bfd..9c1ccc38f3c6a0fbc55c5b352de8fe8bc325641b 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eecfea2f6d4c3644837b80daf0f0625a0db9504e8b3a83272176970c9842edb0
-size 1662985
+oid sha256:33daa078802878e77a250191d319e1a1a450cf38e7a7a1c95c3aec3e57580b2a
+size 2621017
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_FIELD_CLUSTER_METRICS.csv
index 86138990957108e3e6c20338e9aba8f33b167b16..a7c3ac99bd1971bf924056edfb62f48ef6984012 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.6_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5958,41,0.8196302056312561,1.1102656257571446,10,0.6
-inducer,7889,7774,18,0.7151622772216797,2.529341198296715,10,0.6
-strain,2444,2095,30,0.7138633131980896,1.6392458690793867,10,0.6
-expression_strain,5677,5517,41,0.528433620929718,1.9903075044576022,10,0.6
-induction_temperature,3656,3640,13,0.21138402819633484,1.9100818440954646,10,0.6
-source_key,8021,8020,6,0.1000768318772316,3.8781558317316294,10,0.6
-plasmid,5480,5124,122,0.09565138071775436,3.1107746568392542,10,0.6
-molecular_weight,295,277,3,0.08344729989767075,2.3383949078694504,10,0.6
-lysis_buffer,3714,3641,28,0.0711684301495552,2.7114382972997757,10,0.6
-desalting_process,3720,3664,25,0.05286785587668419,2.981808356113664,10,0.6
-enzyme_name,7975,6423,305,0.026753341779112816,3.0762513941146485,10,0.6
-elution_buffer,3530,3440,40,-0.00781943928450346,3.292034369469726,10,0.6
-medium_name,4806,4646,76,-0.06146008521318436,3.0033540728459758,10,0.6
+organism_source,9834,9444,49,0.8377319574356079,1.2884920949306038,10,0.6
+inducer,11954,11851,31,0.6223719120025635,2.338052707440991,10,0.6
+strain,3834,3408,54,0.42171403765678406,2.3010776472613124,10,0.6
+expression_strain,8268,8026,68,0.4137780964374542,2.4872131437978733,10,0.6
+induction_temperature,5383,5348,14,0.23390543460845947,1.7946628666912752,10,0.6
+source_key,12179,12175,9,0.05385942757129669,3.841521041894358,10,0.6
+plasmid,8217,7730,195,0.05150020122528076,3.2021026007570987,10,0.6
+molecular_weight,495,485,5,0.0392894484102726,3.4475670469831905,10,0.6
+desalting_process,5540,5457,35,0.03013639710843563,3.089593743672087,10,0.6
+lysis_buffer,5625,5541,42,0.02949345111846924,2.7793071685760546,10,0.6
+enzyme_name,12099,10318,487,0.014448007568717003,3.108547524519738,10,0.6
+elution_buffer,5198,5085,50,-0.030691243708133698,3.165848036809292,10,0.6
+medium_name,7644,7455,102,-0.07402107864618301,3.215580659008411,10,0.6
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_ALL_FIELDS.csv.gz
index 8489820ecee9112b1414ad5ed8789aa0ede10fd7..21e85e3749042fc2a6faab62967b4a108a574e2b 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89b536b0819fd8d018e09c8386e18c0843678959f2682c52a92174352fefca26
-size 1651983
+oid sha256:d3cf6529420767e6fbcfb19dee3b353001478831d99015f1e56014c753f226a4
+size 2588727
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_FIELD_CLUSTER_METRICS.csv
index 9571709d015106f3458c84a241bd9f0e5846c0c8..9cc27ad0a3dd2b89e82f4fef28e6e4db86a5780c 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.7_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5779,39,0.9292504191398621,0.6710913464361693,10,0.7
-strain,2444,1809,25,0.8110648393630981,1.0468102585669565,10,0.7
-inducer,7889,7694,21,0.6701705455780029,2.0175405256890926,10,0.7
-expression_strain,5677,5192,45,0.5725354552268982,1.0711842577349953,10,0.7
-enzyme_name,7975,3586,161,0.26116514205932617,2.057060465000361,10,0.7
-plasmid,5480,3967,80,0.19424112141132355,2.4261466443782913,10,0.7
-molecular_weight,295,247,11,0.07814138382673264,3.1046419073088063,10,0.7
-medium_name,4806,4167,91,0.04162053391337395,2.700418166736145,10,0.7
-induction_temperature,3656,3591,22,0.005268397741019726,1.9021081432696343,10,0.7
-lysis_buffer,3714,3468,62,0.004350605420768261,2.7925548168920624,10,0.7
-desalting_process,3720,3411,68,-0.00922445673495531,2.814853040631014,10,0.7
-elution_buffer,3530,3209,79,-0.01880348101258278,2.939359933088212,10,0.7
-source_key,8021,7978,79,-0.07267322391271591,2.7770837027808293,10,0.7
+organism_source,9834,9214,44,0.9520369172096252,0.6564892440127331,10,0.7
+inducer,11954,11712,35,0.6442054510116577,2.2726116381470036,10,0.7
+strain,3834,2808,42,0.6395149230957031,1.3463502769382973,10,0.7
+expression_strain,8268,7448,81,0.5286935567855835,1.4613578949112591,10,0.7
+enzyme_name,12099,5718,283,0.2745233476161957,2.029449851798766,10,0.7
+plasmid,8217,5743,137,0.16700774431228638,2.510452225153676,10,0.7
+molecular_weight,495,433,17,0.0645865797996521,3.1048532499101027,10,0.7
+induction_temperature,5383,5302,27,0.04672544449567795,1.9837952912574797,10,0.7
+medium_name,7644,6803,142,0.007070209830999374,2.7109104133113924,10,0.7
+desalting_process,5540,5113,96,-0.022126225754618645,3.035615912164573,10,0.7
+elution_buffer,5198,4763,103,-0.02559037134051323,3.150518424157008,10,0.7
+lysis_buffer,5625,5323,91,-0.029421966522932053,2.9061920150071097,10,0.7
+source_key,12179,12133,101,-0.08255971223115921,3.152832859385902,10,0.7
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_ALL_FIELDS.csv.gz
index e3dec519150a902708e60a1fd7136639cf6bf251..8dfc7f78a39e68b20f76470a91d7c861bb5caa95 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:940c6ff75cabc96443d5d8b0836b8e4da114f28496bc52cab460fbad370057db
-size 1620520
+oid sha256:d330ba6cdfe791afacb25126cb487db838e2b8fd6a780b74d9686ea240decc9c
+size 2538760
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_FIELD_CLUSTER_METRICS.csv
index 2160e0fe163b3e57888790bba5ae9ca63af52925..037737caf8272db706215d6b26693b87cbbed080 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.8_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-strain,2444,1646,17,0.9663563370704651,0.5415234153292803,10,0.8
-organism_source,6292,5628,39,0.9656205177307129,0.4276861324680445,10,0.8
-inducer,7889,7539,37,0.5659745931625366,2.1539705812317784,10,0.8
-expression_strain,5677,5026,47,0.5260725021362305,1.0602940652595259,10,0.8
-induction_temperature,3656,3448,29,0.4183485507965088,1.1540509966722121,10,0.8
-enzyme_name,7975,1715,91,0.3936314880847931,1.5418663237554924,10,0.8
-molecular_weight,295,143,10,0.3780817985534668,1.750687308393917,10,0.8
-medium_name,4806,2963,90,0.2707577347755432,1.9134435561236343,10,0.8
-plasmid,5480,2824,78,0.19203373789787292,1.875099417337347,10,0.8
-desalting_process,3720,2566,95,0.12426748126745224,2.2939916877483464,10,0.8
-elution_buffer,3530,2481,90,0.10220778733491898,2.2268991232358286,10,0.8
-source_key,8021,7474,219,0.08814316242933273,2.349484183878423,10,0.8
-lysis_buffer,3714,2598,88,0.060478437691926956,2.4012693279837634,10,0.8
+organism_source,9834,9034,46,0.962336540222168,0.4490101304115374,10,0.8
+strain,3834,2523,30,0.8005040884017944,0.6760442064711785,10,0.8
+expression_strain,8268,6975,70,0.6330240368843079,1.0370787574631397,10,0.8
+inducer,11954,11442,61,0.5489617586135864,2.0916435470972305,10,0.8
+induction_temperature,5383,5119,37,0.5255248546600342,1.137740434147271,10,0.8
+enzyme_name,12099,2580,145,0.45728006958961487,1.417594959061316,10,0.8
+molecular_weight,495,235,13,0.24558685719966888,1.951701643985915,10,0.8
+medium_name,7644,4893,145,0.2261175960302353,2.0886763350584054,10,0.8
+plasmid,8217,3782,108,0.2081931084394455,1.8660820893530257,10,0.8
+desalting_process,5540,3785,134,0.1046333834528923,2.435494520954159,10,0.8
+elution_buffer,5198,3572,122,0.08296401053667068,2.4036356046347347,10,0.8
+source_key,12179,11494,304,0.052906282246112823,2.521618723112365,10,0.8
+lysis_buffer,5625,4164,145,0.035569578409194946,2.520807883099808,10,0.8
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_ALL_FIELDS.csv.gz
index 9a6fceb905e2e620cdae6e7d217781cdab8b1deb..cc33fb04d1b14e730f267d446a59e101e8a91426 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66ac043c815a4760320bf582a48df6b438af5d08fa9866b2717a8d35fd560fb1
-size 1521113
+oid sha256:67d0ef17cd4bebcbb664ae512cb8364e8610678340919e6700572f5f6a775e29
+size 2359081
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_FIELD_CLUSTER_METRICS.csv
index f90ab78bcd2ceefdadf7018d8da3de23c3d3a08e..2e0eb32cdb958010748745dc4fe08c87cf253724 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=10/t=0.9_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5567,39,0.9992990493774414,0.16990886226485374,10,0.9
-strain,2444,1589,16,0.987883985042572,0.2133845050848035,10,0.9
-induction_temperature,3656,3233,32,0.937194287776947,0.3764394119548351,10,0.9
-enzyme_name,7975,435,28,0.8485289812088013,0.5513011245288655,10,0.9
-molecular_weight,295,23,2,0.8253868818283081,0.47597649720721014,10,0.9
-source_key,8021,1954,136,0.8004114031791687,0.6938050227797024,10,0.9
-desalting_process,3720,650,40,0.7315228581428528,0.8125256731432431,10,0.9
-inducer,7889,7123,84,0.6756018400192261,1.8586509553923383,10,0.9
-expression_strain,5677,4723,62,0.6356610059738159,0.7467334813455969,10,0.9
-plasmid,5480,1938,61,0.6299409866333008,0.7322866208304523,10,0.9
-medium_name,4806,1708,68,0.6277021169662476,0.7732112856895244,10,0.9
-lysis_buffer,3714,769,48,0.605023980140686,1.0599863627776656,10,0.9
-elution_buffer,3530,940,53,0.5420536398887634,1.0813656708608617,10,0.9
+organism_source,9834,8942,44,0.9987958669662476,0.21575083107093854,10,0.9
+induction_temperature,5383,4820,35,0.9458242654800415,0.3651818099487698,10,0.9
+strain,3834,2432,33,0.931756854057312,0.39168613652469375,10,0.9
+enzyme_name,12099,723,48,0.8656179904937744,0.5277958869616057,10,0.9
+molecular_weight,495,54,5,0.864039957523346,0.5792475711619132,10,0.9
+source_key,12179,3354,228,0.7016894221305847,0.8705106968492309,10,0.9
+expression_strain,8268,6457,78,0.6910299062728882,0.6666705373760031,10,0.9
+desalting_process,5540,900,50,0.6737874150276184,0.9744212582553221,10,0.9
+inducer,11954,10782,122,0.6552270650863647,1.9539420014607478,10,0.9
+plasmid,8217,2428,77,0.5940243601799011,0.8175049380900719,10,0.9
+medium_name,7644,2840,92,0.5829232931137085,0.9476279909851705,10,0.9
+lysis_buffer,5625,1255,75,0.5525277256965637,1.116412941291435,10,0.9
+elution_buffer,5198,1347,71,0.48758959770202637,1.2529628557119963,10,0.9
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_ALL_FIELDS.csv.gz
index b12b64b394232ada65c532428f30983d142ec8f6..b659401b51dffd068b62077e03b41bb3703c3f9e 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf5f8b3fe7446fea618b9d0abfd96fe3998b6f41b66afd584f3cbac772f3f005
-size 1638217
+oid sha256:a87a4fd3004c57b4d5aefa1ddfccd3bcabfdd66d8c0091d3a79ae6cd572733d8
+size 2539035
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_FIELD_CLUSTER_METRICS.csv
index a3b619e6b01cd36e23001194521806b90b11d981..25387e8ed15acbcae735e9497ac442a83b3a4134 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.5_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6275,89,0.6493747234344482,1.3335926685862567,2,0.5
-strain,2444,2438,99,0.5401460528373718,1.9845557262841065,2,0.5
-induction_temperature,3656,3656,11,0.381491094827652,1.4979699579079138,2,0.5
-expression_strain,5677,5672,86,0.3620287775993347,1.809875578283524,2,0.5
-inducer,7889,7884,41,0.20281361043453217,1.6490949025736314,2,0.5
-desalting_process,3720,3719,15,0.172710582613945,2.207824584623143,2,0.5
-lysis_buffer,3714,3711,19,0.16232360899448395,1.796967821690144,2,0.5
-elution_buffer,3530,3528,32,0.02386270835995674,2.04887023715694,2,0.5
-plasmid,5480,5476,80,-0.0037308644969016314,2.6889113989081643,2,0.5
-medium_name,4806,4802,92,-0.11628752946853638,2.2527296956809524,2,0.5
-enzyme_name,7975,7963,379,-0.1266491562128067,2.701229793621047,2,0.5
-molecular_weight,295,294,1,,,2,0.5
-source_key,8021,8021,1,,,2,0.5
+organism_source,9834,9820,104,0.7297003269195557,1.333458318078184,2,0.5
+induction_temperature,5383,5383,12,0.40107256174087524,1.7859985847432807,2,0.5
+expression_strain,8268,8266,79,0.3142945468425751,1.845140631539542,2,0.5
+molecular_weight,495,494,2,0.21477341651916504,2.1649878938357836,2,0.5
+inducer,11954,11946,50,0.20274050533771515,2.0211542964564475,2,0.5
+desalting_process,5540,5540,21,0.15065771341323853,2.247119131718312,2,0.5
+lysis_buffer,5625,5623,29,0.14865629374980927,1.7846219127378753,2,0.5
+elution_buffer,5198,5197,45,0.05098419636487961,1.989140371588797,2,0.5
+strain,3834,3824,122,0.04627330228686333,1.9100609494987528,2,0.5
+plasmid,8217,8215,99,-0.015183061361312866,2.953393136891891,2,0.5
+medium_name,7644,7642,118,-0.11563387513160706,2.293491009845395,2,0.5
+enzyme_name,12099,12088,484,-0.13748188316822052,2.7692423473224195,2,0.5
+source_key,12179,12179,1,,,2,0.5
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_ALL_FIELDS.csv.gz
index 888e7bfd64c306f1ac9c54d01667b68e3c200efb..8997d37ec6166cc19ca6c48e0f14f7a1ec7f4367 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6fe6a45ef6a110df3f7c264c9ec22a35171b80550355f700c50278020d2802c7
-size 1702864
+oid sha256:77b9aacccac267ac3eeeb7e97127800e78b74c86dd0ffd08161b8c2637e27e3a
+size 2675469
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_FIELD_CLUSTER_METRICS.csv
index 2e75f81db42719902bbafab79e345dc42cba4332..469f070577387210bf5a1c5a5f446ffd87169f3d 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.6_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6240,122,0.8107720613479614,0.932400669097902,2,0.6
-inducer,7889,7877,73,0.6863388419151306,1.4649668027029443,2,0.6
-strain,2444,2371,129,0.6757716536521912,1.1896439096071307,2,0.6
-expression_strain,5677,5644,108,0.5002840757369995,1.5046906943409648,2,0.6
-induction_temperature,3656,3656,28,0.19832856953144073,1.479254404013885,2,0.6
-plasmid,5480,5423,408,0.0685344710946083,2.05057372229943,2,0.6
-source_key,8021,8021,9,0.05977901071310043,2.958040703463317,2,0.6
-molecular_weight,295,294,9,0.02730863355100155,2.2550583314333705,2,0.6
-lysis_buffer,3714,3705,92,0.024481281638145447,1.7010211864800906,2,0.6
-desalting_process,3720,3714,70,-0.018288983032107353,2.0412683700800227,2,0.6
-enzyme_name,7975,7790,1228,-0.021730097010731697,2.080941473050152,2,0.6
-elution_buffer,3530,3521,117,-0.06796655803918839,2.048690082168262,2,0.6
-medium_name,4806,4773,188,-0.08167680352926254,2.010272599227419,2,0.6
+organism_source,9834,9787,154,0.8251509666442871,1.04265507363664,2,0.6
+inducer,11954,11939,108,0.599524199962616,1.4964930291528231,2,0.6
+strain,3834,3749,215,0.39714863896369934,1.5726498725852744,2,0.6
+expression_strain,8268,8232,200,0.3821985125541687,1.776298981386954,2,0.6
+induction_temperature,5383,5381,37,0.21500450372695923,1.3863549364081367,2,0.6
+source_key,12179,12178,14,0.0033990333322435617,2.911357118344591,2,0.6
+plasmid,8217,8160,677,0.0026394727174192667,2.1431785911739794,2,0.6
+lysis_buffer,5625,5613,124,-0.0095454640686512,1.7468065773677626,2,0.6
+molecular_weight,495,492,12,-0.020108869299292564,2.395293319031555,2,0.6
+desalting_process,5540,5529,100,-0.02550636976957321,2.0281407162615075,2,0.6
+enzyme_name,12099,11878,1667,-0.043349940329790115,2.189702223347721,2,0.6
+elution_buffer,5198,5190,155,-0.08079452812671661,1.9970206486773188,2,0.6
+medium_name,7644,7612,266,-0.1070958822965622,2.0943466750545805,2,0.6
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_ALL_FIELDS.csv.gz
index 9aaa4a9515aacb68eb947d7ee419dfec83315e7f..ad9e278aa18ad089613f5fd4bfef880bbd0b0512 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5cbadf568b7872333f9dc06862e42e71f5e28591876bd0e9ae4b31280d024912
-size 1726340
+oid sha256:c6689091678b0ec7e0de79e39fd2e4d5ef0b4c45379ff2b756b9b329c99acf84
+size 2707761
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_FIELD_CLUSTER_METRICS.csv
index 7407d8690424962f26db30491e348cd6a2c52079..f297b2da4e709c154357c61c966bb630a007e059 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.7_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6218,165,0.9005904197692871,0.5972435173630783,2,0.7
-strain,2444,2301,175,0.7781819701194763,0.8418139012300341,2,0.7
-inducer,7889,7863,90,0.6451825499534607,1.272499603651355,2,0.7
-expression_strain,5677,5577,160,0.5660296678543091,0.9669804583663796,2,0.7
-enzyme_name,7975,6977,1339,0.23361122608184814,1.4868724724350655,2,0.7
-plasmid,5480,5144,541,0.14658498764038086,1.537026517645784,2,0.7
-medium_name,4806,4691,349,0.053585976362228394,1.6486556768801692,2,0.7
-molecular_weight,295,281,30,0.03746595233678818,2.170144910409862,2,0.7
-lysis_buffer,3714,3674,217,-0.023032594472169876,1.6639657487679764,2,0.7
-elution_buffer,3530,3457,238,-0.03465473651885986,1.9646173143057728,2,0.7
-desalting_process,3720,3654,232,-0.03597559407353401,1.8060474007784182,2,0.7
-induction_temperature,3656,3646,50,-0.07302931696176529,1.4475822508585054,2,0.7
-source_key,8021,8019,158,-0.13131871819496155,1.8570237705388797,2,0.7
+organism_source,9834,9751,189,0.9305559992790222,0.7070258047268236,2,0.7
+strain,3834,3570,293,0.6035598516464233,1.0585580188014896,2,0.7
+inducer,11954,11920,131,0.5907979607582092,1.3598098044691096,2,0.7
+expression_strain,8268,8095,311,0.520228922367096,1.1571589668047653,2,0.7
+enzyme_name,12099,10675,2000,0.21953390538692474,1.5377738802721603,2,0.7
+plasmid,8217,7654,904,0.1426750123500824,1.6290487172872608,2,0.7
+molecular_weight,495,481,54,-0.004949654918164015,2.078632474895944,2,0.7
+medium_name,7644,7495,519,-0.007302063517272472,1.6953022017388932,2,0.7
+induction_temperature,5383,5372,62,-0.048322174698114395,1.5362663411348496,2,0.7
+elution_buffer,5198,5096,322,-0.057285554707050323,2.0169840577433322,2,0.7
+desalting_process,5540,5452,333,-0.06141304969787598,1.9341749888426025,2,0.7
+lysis_buffer,5625,5573,313,-0.07052063196897507,1.761737200669923,2,0.7
+source_key,12179,12177,206,-0.12747912108898163,2.0643571615517913,2,0.7
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_ALL_FIELDS.csv.gz
index a4cc953c3a25bc7b3c474f297fe337769e230d50..546ddbe16998c866cc7b3cb32d5e800a87e42620 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2db8aab04920cf701dee8b7960847669121ac03b2b08cb667d198e268f47cd2
-size 1745289
+oid sha256:8aa56d6d2a8a90c0d96e6b06c4b6f99e0cfd2eef3f33aef020dce05fd5afe89b
+size 2732629
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_FIELD_CLUSTER_METRICS.csv
index 9546a38660251c7e6dda56c0f93de375f716e8c7..ad12fa8a44af0a3c626f8cbdea2c2d49092db969 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.8_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6179,199,0.9589523673057556,0.4259569163460959,2,0.8
-strain,2444,2236,198,0.9366304278373718,0.5265192971144985,2,0.8
-inducer,7889,7811,148,0.5388298034667969,1.1960512931785707,2,0.8
-expression_strain,5677,5503,188,0.5094572305679321,0.7518032089201473,2,0.8
-enzyme_name,7975,5663,1421,0.43413490056991577,1.0658222506577908,2,0.8
-induction_temperature,3656,3622,84,0.39889413118362427,0.9730751699724313,2,0.8
-plasmid,5480,4620,696,0.30422961711883545,1.099124661903656,2,0.8
-molecular_weight,295,249,48,0.3012656271457672,1.373393176516414,2,0.8
-medium_name,4806,4356,588,0.28813114762306213,1.1767819115561158,2,0.8
-desalting_process,3720,3384,443,0.1946047842502594,1.4288960943955102,2,0.8
-elution_buffer,3530,3183,398,0.15043626725673676,1.4326258881485796,2,0.8
-lysis_buffer,3714,3466,427,0.12739808857440948,1.4573947029247922,2,0.8
-source_key,8021,7981,508,0.0952213704586029,1.5720090407930014,2,0.8
+organism_source,9834,9699,238,0.9578798413276672,0.4295251319180712,2,0.8
+strain,3834,3403,314,0.7935447096824646,0.5645557212239304,2,0.8
+expression_strain,8268,7893,357,0.6261911392211914,0.7387327043474097,2,0.8
+inducer,11954,11852,234,0.5297640562057495,1.281111716423976,2,0.8
+induction_temperature,5383,5337,108,0.4958697259426117,1.0970403121093641,2,0.8
+enzyme_name,12099,8546,2137,0.46323251724243164,1.0307130690496105,2,0.8
+plasmid,8217,6653,1107,0.32409244775772095,1.1258290764486218,2,0.8
+medium_name,7644,6907,892,0.2726289629936218,1.2701827177499012,2,0.8
+molecular_weight,495,427,79,0.2031012326478958,1.451036215754577,2,0.8
+desalting_process,5540,5004,655,0.1561461091041565,1.4810227851066644,2,0.8
+elution_buffer,5198,4693,610,0.13465425372123718,1.4870304336039033,2,0.8
+lysis_buffer,5625,5280,622,0.09519679844379425,1.5096713079035113,2,0.8
+source_key,12179,12134,748,0.041748207062482834,1.643826991204593,2,0.8
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_ALL_FIELDS.csv.gz
index fb3ad910302d72f93e6134ab2bbac2dcb30f7fd4..6cb972d0679df3dd342d6be127b5d34ae725daac 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad0c6c7aa5d3fb6d4caed870011794a591459d20d7cc2783efa435a94b1fa8f6
-size 1676471
+oid sha256:5fea14a17de5f5c07a9bef2709f0364d4dc401dd883bb35fddff13fcf8023a66
+size 2602062
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_FIELD_CLUSTER_METRICS.csv
index 7715c00133bcabcaa88c120e85b0cc5aa40eccc9..c0402d1fead80313dc20c8f43f2a2beab8805f7f 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=2/t=0.9_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6155,213,0.9978790879249573,0.2220522197286095,2,0.9
-strain,2444,2182,207,0.9788490533828735,0.26372168970471044,2,0.9
-induction_temperature,3656,3562,130,0.9164313077926636,0.3503126642099208,2,0.9
-source_key,8021,7422,1543,0.869504451751709,0.5365352651473726,2,0.9
-desalting_process,3720,2635,691,0.8025253415107727,0.5946045758883378,2,0.9
-molecular_weight,295,192,49,0.796576976776123,0.5855528611347344,2,0.9
-enzyme_name,7975,3547,1127,0.7588869333267212,0.59197278180923,2,0.9
-lysis_buffer,3714,2885,720,0.7397023439407349,0.6566168869121246,2,0.9
-medium_name,4806,3784,729,0.7169615030288696,0.5280011942683516,2,0.9
-plasmid,5480,3840,716,0.6763919591903687,0.5589648741586412,2,0.9
-elution_buffer,3530,2596,579,0.6561211943626404,0.6875667523404637,2,0.9
-expression_strain,5677,5422,269,0.6520324945449829,0.5207156191048923,2,0.9
-inducer,7889,7696,323,0.6496630311012268,1.1189665679031124,2,0.9
+organism_source,9834,9659,255,0.9977973699569702,0.22539752638533583,2,0.9
+induction_temperature,5383,5255,162,0.9204808473587036,0.4174417727951194,2,0.9
+strain,3834,3284,318,0.9085197448730469,0.31850613964948793,2,0.9
+source_key,12179,11260,2293,0.8043889999389648,0.6058066734653524,2,0.9
+molecular_weight,495,320,81,0.8018842935562134,0.550827475788505,2,0.9
+enzyme_name,12099,5445,1726,0.7885855436325073,0.5508262686704151,2,0.9
+desalting_process,5540,3785,1021,0.773786187171936,0.6372258062331936,2,0.9
+lysis_buffer,5625,4334,1069,0.7125115990638733,0.6797550928998537,2,0.9
+expression_strain,8268,7715,452,0.7048622965812683,0.4681556602121261,2,0.9
+medium_name,7644,5945,1091,0.6938697695732117,0.5468883421123496,2,0.9
+plasmid,8217,5312,1089,0.6720221638679504,0.5692584671385185,2,0.9
+elution_buffer,5198,3690,861,0.6556960940361023,0.6907750388673599,2,0.9
+inducer,11954,11639,490,0.630872905254364,1.1369147053926885,2,0.9
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_ALL_FIELDS.csv.gz
index 78da0d17d580587e46d46fb69887a90b3adf27bf..edc3408c9eee3ac5715949b1fde739161d6327df 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3eae8a6ab6064073139f7d7c09370123e4c18e5a557efe6e6939abd20cc19afb
-size 1619338
+oid sha256:c2ef7c9cdb57956194fcfe565e4b980d126df9ece4f8b62bebde74ca42599813
+size 2517331
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_FIELD_CLUSTER_METRICS.csv
index 41a5360d4021f3522d60839a26892b6465e08762..9a532e5ff8af94b4abdbc204ffc8434edc06a224 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.5_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6123,28,0.6570871472358704,1.712664195480983,20,0.5
-strain,2444,2289,21,0.5660344958305359,3.172909053765197,20,0.5
-induction_temperature,3656,3629,4,0.449496865272522,1.6747973500449849,20,0.5
-expression_strain,5677,5606,20,0.38536372780799866,2.8329020679723307,20,0.5
-inducer,7889,7832,7,0.3055448830127716,2.572608839809441,20,0.5
-lysis_buffer,3714,3698,6,0.2301340103149414,3.1684643062150832,20,0.5
-desalting_process,3720,3686,4,0.22366729378700256,3.004260552478986,20,0.5
-elution_buffer,3530,3482,10,0.09006357938051224,3.405307240355757,20,0.5
-plasmid,5480,5392,22,0.046792831271886826,4.5591082114810115,20,0.5
-medium_name,4806,4732,27,-0.021849416196346283,3.8265609314767515,20,0.5
-enzyme_name,7975,7686,102,-0.04246228188276291,4.4339251069130174,20,0.5
-molecular_weight,295,294,1,,,20,0.5
-source_key,8021,8021,1,,,20,0.5
+organism_source,9834,9630,29,0.74268639087677,1.8294043566657425,20,0.5
+induction_temperature,5383,5355,5,0.44859176874160767,1.7206507581173347,20,0.5
+expression_strain,8268,8175,21,0.3155079483985901,2.96855202464295,20,0.5
+inducer,11954,11871,12,0.2602851986885071,3.6721095969909308,20,0.5
+lysis_buffer,5625,5585,7,0.2146769016981125,3.2321077126832383,20,0.5
+desalting_process,5540,5520,6,0.20954683423042297,3.7475563284586038,20,0.5
+strain,3834,3700,26,0.14651739597320557,3.4702286695897677,20,0.5
+elution_buffer,5198,5149,12,0.09627825766801834,3.2127838851151487,20,0.5
+plasmid,8217,8137,30,0.04867859557271004,5.094660580475175,20,0.5
+medium_name,7644,7565,36,-0.04143005982041359,3.973477521825681,20,0.5
+enzyme_name,12099,11804,138,-0.041651807725429535,4.529566985073866,20,0.5
+molecular_weight,495,491,1,,,20,0.5
+source_key,12179,12179,1,,,20,0.5
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_ALL_FIELDS.csv.gz
index dea61a01adbc6f10009293499dab8daaa1ba2e20..c78d5c28b3606b6c0768889229df50ff484a6fc6 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bdabcfa35fa077b1ab8b0c9bc2e47ecd8c7d054ec86b2e79fa0daf1d3974de44
-size 1631893
+oid sha256:e6dc1215e58e248fecb1e6b14fe5e69753d5c74094babfacd04617c5f6b3c9c2
+size 2569919
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_FIELD_CLUSTER_METRICS.csv
index cec1b38b1fe237b51d87e5a1647e6cd688c81903..47f8a94829a94793316cee126ee95c9d125fbc52 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.6_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5753,24,0.8317749500274658,1.0055797635872468,20,0.6
-strain,2444,1863,11,0.7663013339042664,1.778019325959417,20,0.6
-inducer,7889,7724,10,0.7234408855438232,2.7315032531394485,20,0.6
-expression_strain,5677,5380,23,0.5388358235359192,1.9810552935499508,20,0.6
-induction_temperature,3656,3632,9,0.24118949472904205,2.129030426920334,20,0.6
-plasmid,5480,4861,62,0.11633288115262985,3.442479747045062,20,0.6
-molecular_weight,295,273,2,0.11602164059877396,3.361375652776792,20,0.6
-source_key,8021,8017,5,0.1030571386218071,4.061771773391138,20,0.6
-lysis_buffer,3714,3630,18,0.08762457221746445,2.9127973684576363,20,0.6
-desalting_process,3720,3609,15,0.07537908107042313,3.179134294776404,20,0.6
-enzyme_name,7975,5359,141,0.04196884110569954,3.414234715405289,20,0.6
-elution_buffer,3530,3390,27,-9.898182906908914e-05,3.7053614142870486,20,0.6
-medium_name,4806,4532,46,-0.04943051561713219,3.5758910500147163,20,0.6
+organism_source,9834,9188,28,0.8608959913253784,1.2135624267790803,20,0.6
+inducer,11954,11778,16,0.6450905799865723,2.728055524697932,20,0.6
+strain,3834,3073,22,0.4684427082538605,2.404375209550413,20,0.6
+expression_strain,8268,7908,45,0.42838072776794434,2.6549104971211714,20,0.6
+induction_temperature,5383,5349,11,0.26610639691352844,1.9256324125623323,20,0.6
+molecular_weight,495,474,3,0.0906362384557724,3.650137102103802,20,0.6
+lysis_buffer,5625,5509,27,0.08258143812417984,3.2625219767676716,20,0.6
+source_key,12179,12169,7,0.05876987427473068,4.330825269550882,20,0.6
+plasmid,8217,7345,110,0.05808311700820923,3.6218252627404333,20,0.6
+desalting_process,5540,5426,22,0.049398571252822876,3.219527555698754,20,0.6
+enzyme_name,12099,8721,233,0.030406733974814415,3.4452921913839485,20,0.6
+elution_buffer,5198,4984,27,-0.00635951291769743,3.6204982115714643,20,0.6
+medium_name,7644,7311,62,-0.06386596709489822,3.731007632587912,20,0.6
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_ALL_FIELDS.csv.gz
index c06b801fb44f53832923135ea413bc3bc87330ba..fd12718834e342ec8539e46cf4da4c6ea53d9af3 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce6097ecfa47481228a98a76b9f44da7b3a6e73941be9ffe2f1039952e0d075d
-size 1615488
+oid sha256:cc3002bc52d6a254b5a52b9743747b1b0f57fe9ff9b2e4c09043b10e50f03685
+size 2519800
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_FIELD_CLUSTER_METRICS.csv
index bb408700ec43f780b4391fe9f8d0d4f292d0e221..ebb20873e68c5f97f900b21d62d8a80122266493 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.7_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5520,20,0.942322850227356,0.5266400513991684,20,0.7
-strain,2444,1583,7,0.8380798697471619,0.9931627812905172,20,0.7
-inducer,7889,7612,10,0.6981210112571716,1.900579675497409,20,0.7
-expression_strain,5677,4997,26,0.5969026684761047,1.0494110044701601,20,0.7
-enzyme_name,7975,2274,50,0.262971431016922,2.2775209433027177,20,0.7
-plasmid,5480,3586,38,0.21920683979988098,2.636374044476015,20,0.7
-molecular_weight,295,229,6,0.12021034210920334,3.2785726459361633,20,0.7
-medium_name,4806,3808,49,0.04216783866286278,3.147005943408512,20,0.7
-induction_temperature,3656,3556,14,0.03346134349703789,2.053274227061542,20,0.7
-lysis_buffer,3714,3318,34,0.01112122181802988,3.3312675251319135,20,0.7
-desalting_process,3720,3255,36,0.0063720569014549255,3.2584231364610172,20,0.7
-elution_buffer,3530,3079,49,-0.011515987105667591,3.259201634314855,20,0.7
-source_key,8021,7954,54,-0.052339281886816025,3.3779526592817612,20,0.7
+organism_source,9834,8931,24,0.9596486687660217,0.6216337303586462,20,0.7
+strain,3834,2520,19,0.668351411819458,1.3244934409755913,20,0.7
+inducer,11954,11597,19,0.6515691876411438,2.5134688564067402,20,0.7
+expression_strain,8268,7086,44,0.5501809120178223,1.221030917855851,20,0.7
+enzyme_name,12099,3489,87,0.29934900999069214,2.1161257479472555,20,0.7
+plasmid,8217,5010,62,0.19646042585372925,2.709583243823644,20,0.7
+molecular_weight,495,389,9,0.08801067620515823,3.2699986829618344,20,0.7
+induction_temperature,5383,5284,20,0.051803335547447205,2.136933130847926,20,0.7
+medium_name,7644,6294,69,0.012482562102377415,3.0980979395002284,20,0.7
+desalting_process,5540,4951,59,-0.0006680836086161435,3.3401313132624577,20,0.7
+elution_buffer,5198,4548,58,-0.010796690359711647,3.7180590968970724,20,0.7
+lysis_buffer,5625,5054,49,-0.019271939992904663,3.329629282532879,20,0.7
+source_key,12179,12069,72,-0.058189522475004196,3.6488871353345584,20,0.7
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_ALL_FIELDS.csv.gz
index 632b3f3993243536b5dea2dbe2bae43e195fc8a9..fe72c2fe2baa3b31476a2229499357eed310c324 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d8febf9b431fb0ec53124b0e3ea86407b0d0ba3059f9b49c736da1e196e4a5e
-size 1575449
+oid sha256:a9a503930f44ec352058dfa3003d70cc79c54d6104433323d467dcbfbeafd86c
+size 2454661
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_FIELD_CLUSTER_METRICS.csv
index 253a8c202ed927756c8a8231ddf3a60e2864afd2..adcffa7c065a2b43db7aab6a5db0d1d5ad04e058 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.8_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-strain,2444,1501,6,0.9720566868782043,0.5801222342703348,20,0.8
-organism_source,6292,5346,18,0.9681975841522217,0.06309666212237781,20,0.8
-inducer,7889,7434,22,0.5859264731407166,2.6714871494335597,20,0.8
-expression_strain,5677,4773,26,0.5387418866157532,1.1143419361252411,20,0.8
-induction_temperature,3656,3362,21,0.4482060968875885,1.194106525289084,20,0.8
-molecular_weight,295,50,2,0.3570627272129059,1.713085226764326,20,0.8
-enzyme_name,7975,982,25,0.32650333642959595,1.9047709904999752,20,0.8
-medium_name,4806,2442,41,0.2748943865299225,2.136216130049567,20,0.8
-plasmid,5480,2386,32,0.20637382566928864,2.1844799421946015,20,0.8
-desalting_process,3720,2075,40,0.11954550445079803,2.797215181163522,20,0.8
-source_key,8021,6932,129,0.09655717760324478,2.680824601823045,20,0.8
-elution_buffer,3530,2080,42,0.06262142211198807,2.7391748808439695,20,0.8
-lysis_buffer,3714,2164,42,0.038235198706388474,2.8724469998404727,20,0.8
+organism_source,9834,8717,22,0.965220034122467,0.26762438650056414,20,0.8
+strain,3834,2348,16,0.7916683554649353,0.8387718984847852,20,0.8
+expression_strain,8268,6607,40,0.6468216776847839,1.053406578171578,20,0.8
+inducer,11954,11256,35,0.5667756795883179,2.5790076380515936,20,0.8
+induction_temperature,5383,4975,24,0.5292564630508423,1.0757273567047425,20,0.8
+enzyme_name,12099,1290,36,0.43254441022872925,1.5407455347789045,20,0.8
+medium_name,7644,4171,69,0.24280966818332672,2.327841548549386,20,0.8
+molecular_weight,495,140,5,0.21755020320415497,2.1694918521617295,20,0.8
+plasmid,8217,3212,50,0.20771467685699463,2.0972788126486526,20,0.8
+desalting_process,5540,3118,61,0.09698440134525299,2.8289141469602836,20,0.8
+elution_buffer,5198,3046,57,0.06896746158599854,2.6989111846321183,20,0.8
+source_key,12179,10745,184,0.06556087732315063,2.853161131704382,20,0.8
+lysis_buffer,5625,3478,67,0.03756510838866234,2.8761167602475366,20,0.8
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_ALL_FIELDS.csv.gz
index 6a15d48a08f621b28c0b6d6d45528db856ad180c..4789d88e04ecd1c796335eaa90aac744e35d16de 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1db3e884f1d5c68d7ccb93f5c81e7a873b1c43e9dc3feb1bb3b2304fe69dc85f
-size 1490709
+oid sha256:fbaea43ceb356b07cd988c57c4c2d76b683b47a2197bbcccf933dc3e9a2a9601
+size 2312310
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_FIELD_CLUSTER_METRICS.csv
index 89a1f402101fa5b106369a9e0b099aa6bcb75299..6766d97404c42ee7f5c581dfaef6a9ed88feffac 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=20/t=0.9_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5313,19,0.9997736215591431,0.0065828248116430385,20,0.9
-strain,2444,1438,4,0.9890567660331726,0.18774175555253358,20,0.9
-induction_temperature,3656,3127,24,0.9370285868644714,0.3899088425412159,20,0.9
-enzyme_name,7975,102,3,0.9331685900688171,0.3102416133624983,20,0.9
-desalting_process,3720,280,8,0.7853027582168579,0.7971958726644104,20,0.9
-source_key,8021,449,19,0.7609233856201172,0.7663704282215069,20,0.9
-inducer,7889,6805,46,0.708914041519165,1.8338011885467183,20,0.9
-medium_name,4806,1038,14,0.704494059085846,0.7417135074533199,20,0.9
-expression_strain,5677,4399,34,0.6388029456138611,0.8087281217255706,20,0.9
-plasmid,5480,1547,27,0.6263361573219299,0.8726016662074776,20,0.9
-lysis_buffer,3714,305,10,0.5956308841705322,0.9948648097772527,20,0.9
-elution_buffer,3530,440,13,0.49330535531044006,1.2777328463171371,20,0.9
-molecular_weight,295,0,0,,,20,0.9
+organism_source,9834,8658,22,0.999137818813324,0.21601091402703407,20,0.9
+induction_temperature,5383,4700,26,0.94735187292099,0.3712563574186719,20,0.9
+enzyme_name,12099,134,4,0.9376413226127625,0.2932256696553328,20,0.9
+strain,3834,2205,15,0.9336280822753906,0.4347388338169246,20,0.9
+desalting_process,5540,413,9,0.7524581551551819,0.7920455369776906,20,0.9
+expression_strain,8268,6063,46,0.6981281638145447,0.5331694307735217,20,0.9
+source_key,12179,866,34,0.67783123254776,0.9420713594528795,20,0.9
+inducer,11954,10479,75,0.6695730090141296,2.171065292254777,20,0.9
+plasmid,8217,1846,30,0.6068649291992188,0.9391116226911075,20,0.9
+medium_name,7644,1961,23,0.6012668013572693,1.0455487201580906,20,0.9
+elution_buffer,5198,787,23,0.4685429334640503,1.4413275362887834,20,0.9
+lysis_buffer,5625,586,19,0.465267151594162,1.3649728216905517,20,0.9
+molecular_weight,495,0,0,,,20,0.9
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_ALL_FIELDS.csv.gz
index 0736702e41def5c194b230de6eba277018c13d61..9831601e4dc412a5ff904e63bb02ef60c1c97608 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00b3820663b6fff7bec84fee84a73942e72b5412edf52d9015492470f25a60a0
-size 1610009
+oid sha256:5dcb06ef879ab0a1e53cb7bac2e43ef1421135af198cab7d8c3385dd3a867228
+size 2503886
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_FIELD_CLUSTER_METRICS.csv
index fdcd1da06d0c1ff3647aae84d9fbee3aeae2f343..d7d4a7b31dd075d8132f046a4cefe25272b97698 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.5_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6008,21,0.6594285368919373,1.666366840800708,30,0.5
-strain,2444,2199,15,0.5773423314094543,3.3662626954023396,30,0.5
-induction_temperature,3656,3638,4,0.45472609996795654,1.7889025435898822,30,0.5
-expression_strain,5677,5575,13,0.40181440114974976,2.817033917536209,30,0.5
-inducer,7889,7793,5,0.3097394108772278,2.803125781978495,30,0.5
-desalting_process,3720,3686,3,0.24632686376571655,2.7690726772977254,30,0.5
-lysis_buffer,3714,3680,4,0.23621866106987,3.0684415814247465,30,0.5
-elution_buffer,3530,3469,7,0.0986451804637909,3.90376472208463,30,0.5
-plasmid,5480,5370,16,0.06411071866750717,5.221126692039306,30,0.5
-medium_name,4806,4692,22,-0.020979417487978935,3.9243444209606433,30,0.5
-enzyme_name,7975,7571,73,-0.03500885143876076,4.772294920257217,30,0.5
-molecular_weight,295,294,1,,,30,0.5
-source_key,8021,8021,1,,,30,0.5
+organism_source,9834,9542,23,0.7454800009727478,1.8523417284342252,30,0.5
+induction_temperature,5383,5372,5,0.45750507712364197,1.8246663219356425,30,0.5
+expression_strain,8268,8179,16,0.32264626026153564,3.3043814048393507,30,0.5
+inducer,11954,11834,8,0.2734537422657013,3.71851802615069,30,0.5
+desalting_process,5540,5488,4,0.2227264642715454,3.9897919816594425,30,0.5
+lysis_buffer,5625,5559,5,0.22032476961612701,3.4059473053285756,30,0.5
+strain,3834,3586,17,0.15691837668418884,3.8353865977395447,30,0.5
+elution_buffer,5198,5117,9,0.10266560316085815,3.193460167871953,30,0.5
+plasmid,8217,8093,25,0.051208510994911194,5.294839074109648,30,0.5
+medium_name,7644,7507,26,-0.031139198690652847,4.208441617103189,30,0.5
+enzyme_name,12099,11681,108,-0.03540992736816406,4.750936676894225,30,0.5
+molecular_weight,495,491,1,,,30,0.5
+source_key,12179,12179,1,,,30,0.5
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_ALL_FIELDS.csv.gz
index 8d2460b893b00512216aecb60ce37188a94ea90a..803efadc54d7858aad581dbd9864f3a98df6fbb4 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b544c56a3d5a954e806ac0ab868294f3be5c30d0e253ece0ca63a7bed213215
-size 1612566
+oid sha256:ec252aa1932dbbe2c85a519829a2d06a46ab612789a91a0c87291597ff3e30b3
+size 2536015
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_FIELD_CLUSTER_METRICS.csv
index 4cc81d664a8f22d4da70d9f2326b215f240aea44..9b3eee01239efff927864544c2c4014a4720012d 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.6_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5612,17,0.8399215340614319,0.8450379627914887,30,0.6
-strain,2444,1767,7,0.8014580607414246,1.4780306540745123,30,0.6
-inducer,7889,7688,7,0.7296125888824463,2.2042269872974054,30,0.6
-expression_strain,5677,5301,17,0.5527005791664124,1.9502435961774685,30,0.6
-induction_temperature,3656,3615,7,0.3207237124443054,1.8903240241138224,30,0.6
-plasmid,5480,4662,41,0.1289088875055313,3.509551727572814,30,0.6
-source_key,8021,7997,4,0.10551067441701889,4.284743208245457,30,0.6
-lysis_buffer,3714,3566,12,0.09923949837684631,3.1585074416780574,30,0.6
-desalting_process,3720,3541,10,0.0864974856376648,3.233871115438056,30,0.6
-enzyme_name,7975,4608,81,0.039102230221033096,3.654164090511761,30,0.6
-elution_buffer,3530,3306,17,0.021820921450853348,3.886848114537697,30,0.6
-medium_name,4806,4479,36,-0.02778339944779873,3.7079294312820363,30,0.6
-molecular_weight,295,247,1,,,30,0.6
+organism_source,9834,8915,16,0.8754450678825378,1.0016939691793685,30,0.6
+inducer,11954,11706,10,0.651712954044342,2.4326283038571637,30,0.6
+strain,3834,2979,17,0.48042282462120056,2.428655958398597,30,0.6
+expression_strain,8268,7792,34,0.43849411606788635,2.865285877619293,30,0.6
+induction_temperature,5383,5333,8,0.34988197684288025,2.0646286948036097,30,0.6
+molecular_weight,495,447,2,0.10484033823013306,3.9815982575491367,30,0.6
+lysis_buffer,5625,5423,18,0.08649435639381409,3.8625711883072222,30,0.6
+desalting_process,5540,5351,15,0.07188691198825836,3.501662729536427,30,0.6
+source_key,12179,12141,5,0.06901775300502777,5.223996047470979,30,0.6
+plasmid,8217,7065,77,0.06764841824769974,3.8300421736200696,30,0.6
+enzyme_name,12099,7649,142,0.041884299367666245,3.566055129031407,30,0.6
+elution_buffer,5198,4916,20,0.01682133786380291,3.6571588571743563,30,0.6
+medium_name,7644,7228,49,-0.06289418786764145,3.970248521320083,30,0.6
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_ALL_FIELDS.csv.gz
index 693f67c623136f594110a4818f3056523ab0e374..888b9af7796e660c81900d4573950f4643bfe179 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c8b4f4a28fba1ad4630a60181305269137bc60b244ac1e3376ed36cd6d8af23
-size 1590907
+oid sha256:5c60b39b50afac8abd539825ca2dd12cd93d8568e179a470c6c6e5bd1fb19194
+size 2482472
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_FIELD_CLUSTER_METRICS.csv
index 5fb0f9e0b1eab1b1da0fae70a8b24d71a7dd0cb0..9a8155362934f6fd07b7b8f1cc2c694ba639391a 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.7_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5368,14,0.944336473941803,0.5096058626872298,30,0.7
-strain,2444,1509,4,0.8467665910720825,1.0025612881676649,30,0.7
-inducer,7889,7565,7,0.7035017013549805,1.8093146282427004,30,0.7
-expression_strain,5677,4867,18,0.6164101958274841,1.0858407562011232,30,0.7
-enzyme_name,7975,1749,23,0.25996050238609314,2.246177216071364,30,0.7
-plasmid,5480,3261,21,0.21507234871387482,3.011486811827421,30,0.7
-molecular_weight,295,179,3,0.13212351500988007,3.551529893466173,30,0.7
-induction_temperature,3656,3527,11,0.12676669657230377,1.9583043360024748,30,0.7
-medium_name,4806,3612,34,0.059124723076820374,3.142537266903016,30,0.7
-desalting_process,3720,3130,24,0.015858545899391174,3.5602085609598553,30,0.7
-lysis_buffer,3714,3102,20,0.010213813744485378,3.82921337040075,30,0.7
-elution_buffer,3530,2942,33,-0.007943201810121536,3.519595296054964,30,0.7
-source_key,8021,7903,41,-0.03699294850230217,3.704612828261104,30,0.7
+organism_source,9834,8785,18,0.9626603126525879,0.546461446965384,30,0.7
+strain,3834,2365,12,0.6824359893798828,1.2107708727400417,30,0.7
+inducer,11954,11538,12,0.6760551929473877,2.5446635402907294,30,0.7
+expression_strain,8268,6823,29,0.5677508115768433,1.1975121173914587,30,0.7
+enzyme_name,12099,2570,42,0.3194103240966797,1.9730449965561097,30,0.7
+plasmid,8217,4459,33,0.20662137866020203,2.6907863309572786,30,0.7
+molecular_weight,495,330,5,0.12120842188596725,2.926746176314895,30,0.7
+induction_temperature,5383,5207,13,0.10135113447904587,2.065552348101593,30,0.7
+medium_name,7644,6009,49,0.03147301822900772,3.3560253115098915,30,0.7
+desalting_process,5540,4771,40,0.005013736896216869,3.6910876778502653,30,0.7
+elution_buffer,5198,4438,40,0.0017843650421127677,3.858868515542658,30,0.7
+lysis_buffer,5625,4897,34,-0.012063998728990555,3.6496424811196446,30,0.7
+source_key,12179,12025,54,-0.0484514981508255,4.113655225220298,30,0.7
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_ALL_FIELDS.csv.gz
index 0542bbe57044aa257f7db224ad8caa85eb2635c5..c05fe9017a0ec379fa95b4999cbcaa299a3d8657 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:735b69fcf561abc5a75a5fa09646838f5c7db31eb77ebc93ffe3159e866cd77c
-size 1549037
+oid sha256:2df30662206aa0ebf0fb5b12b0fa2c442710d0aa4aba1cd028a598d9f5c10aff
+size 2417442
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_FIELD_CLUSTER_METRICS.csv
index 3758699ec2c6c88ce75afa677de530876f025f2a..b16b2d1a8e9022f82874992dfd9198951e1664c6 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.8_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-strain,2444,1427,3,0.9845060706138611,0.2565503320277799,30,0.8
-organism_source,6292,5219,13,0.9674237370491028,0.06670955409641507,30,0.8
-inducer,7889,7394,18,0.5881363749504089,2.8128090150048535,30,0.8
-expression_strain,5677,4610,18,0.5393204092979431,1.1186158541313438,30,0.8
-induction_temperature,3656,3282,17,0.46030616760253906,1.1818300064423306,30,0.8
-enzyme_name,7975,676,11,0.38152819871902466,1.7385168773185329,30,0.8
-medium_name,4806,2144,25,0.3440277874469757,2.162077071336094,30,0.8
-plasmid,5480,2170,21,0.190673828125,2.2474881160957336,30,0.8
-desalting_process,3720,1842,25,0.146604984998703,2.8367336595982637,30,0.8
-source_key,8021,6351,91,0.10296914726495743,2.8326877864557813,30,0.8
-elution_buffer,3530,1814,26,0.07680466026067734,2.697036831603645,30,0.8
-lysis_buffer,3714,1898,26,0.03304087370634079,3.181086753288449,30,0.8
-molecular_weight,295,30,1,,,30,0.8
+organism_source,9834,8586,17,0.9658926725387573,0.1661358028135214,30,0.8
+strain,3834,2248,12,0.789456307888031,0.9267350094374822,30,0.8
+expression_strain,8268,6241,23,0.7027596235275269,0.8462549967929675,30,0.8
+inducer,11954,11169,26,0.5698144435882568,2.765250785084089,30,0.8
+induction_temperature,5383,4893,21,0.531478226184845,1.0642951032000039,30,0.8
+enzyme_name,12099,846,16,0.4489150941371918,1.4720061319940563,30,0.8
+medium_name,7644,3729,41,0.2876500189304352,2.192163531256912,30,0.8
+plasmid,8217,2817,29,0.18388640880584717,2.4662787989475885,30,0.8
+desalting_process,5540,2861,40,0.10724735260009766,2.9026030394286946,30,0.8
+elution_buffer,5198,2707,36,0.08045176416635513,2.907047851920069,30,0.8
+source_key,12179,10049,127,0.06502120196819305,3.100974361843955,30,0.8
+lysis_buffer,5625,3165,45,0.042675722390413284,3.0614064035212745,30,0.8
+molecular_weight,495,55,1,,,30,0.8
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_ALL_FIELDS.csv.gz
index 27c3c003535cd68eebaa394032c5d7a3285c0750..465c363716369c10872a3638c3233530b09a969b 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2190cb54b0980e74894b7b0c6a243c49cac7f483ce56c392eaf5e4b1d2af5687
-size 1481808
+oid sha256:22b71eaa578743da490376a3ab391293db70cf387a15ed6ff4f81c67aa7cffdf
+size 2296520
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_FIELD_CLUSTER_METRICS.csv
index 5520ef257109f1d4fdaaa7837cff7b803ce2772f..8d9089ca1dd34f40cc9f14050575a49b3e1d580f 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=30/t=0.9_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5186,14,0.9997681975364685,0.006938165148779805,30,0.9
-strain,2444,1415,3,0.9917000532150269,0.18818972104813725,30,0.9
-induction_temperature,3656,3007,19,0.9491093158721924,0.33276716686109964,30,0.9
-enzyme_name,7975,76,2,0.9301405549049377,0.3334401254655186,30,0.9
-desalting_process,3720,186,4,0.8694847822189331,0.5047766294503007,30,0.9
-source_key,8021,63,2,0.8231516480445862,0.6109451022231208,30,0.9
-inducer,7889,6496,28,0.7248172163963318,1.9890572791448466,30,0.9
-medium_name,4806,951,10,0.7143369913101196,0.8118360327693084,30,0.9
-plasmid,5480,1272,16,0.6723026037216187,0.8419289729023927,30,0.9
-expression_strain,5677,4187,22,0.6518675684928894,0.6891774604763982,30,0.9
-elution_buffer,3530,257,5,0.5735747218132019,1.1429962703010648,30,0.9
-lysis_buffer,3714,198,5,0.5676564574241638,1.097927471279638,30,0.9
-molecular_weight,295,0,0,,,30,0.9
+organism_source,9834,8523,17,0.999859094619751,0.007527305537962825,30,0.9
+induction_temperature,5383,4548,20,0.9544343948364258,0.3465218882129265,30,0.9
+strain,3834,2132,12,0.9349521994590759,0.44326260171728044,30,0.9
+enzyme_name,12099,111,3,0.9257763624191284,0.33107063290330735,30,0.9
+desalting_process,5540,345,6,0.800753116607666,0.7057760803742767,30,0.9
+expression_strain,8268,5695,30,0.7017275094985962,0.5984569948395705,30,0.9
+inducer,11954,10095,44,0.6953803896903992,2.2599839630977643,30,0.9
+source_key,12179,230,6,0.6611170172691345,0.9629723096666277,30,0.9
+plasmid,8217,1571,19,0.6399921178817749,0.9679614343462944,30,0.9
+medium_name,7644,1776,15,0.6185808181762695,0.9203057075402886,30,0.9
+lysis_buffer,5625,308,7,0.5914812684059143,1.112702945092193,30,0.9
+elution_buffer,5198,536,11,0.49873489141464233,1.4797777069838673,30,0.9
+molecular_weight,495,0,0,,,30,0.9
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_ALL_FIELDS.csv.gz
index 06d06f124e39b4061a6fb4fc70e43eeaac06092f..36c32665509c4545237143ee5ac3002971c093dd 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ca46f0c63683b3559a775f294c74a4168fe6b7969efa39792d0aa9760b7006f
-size 1635441
+oid sha256:710c0132cdd9f2d7021e83fcee1a74e811f32d43aeee806359aab7fd6cb55017
+size 2536814
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_FIELD_CLUSTER_METRICS.csv
index 893a72747cd06ce0679c737e7eee0ac57cf031fe..b80c20853b59d23383ec270c0e6b8ce507e82234 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.5_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6237,56,0.6501827836036682,1.6181314594087712,5,0.5
-strain,2444,2401,57,0.5443737506866455,2.515341602080898,5,0.5
-induction_temperature,3656,3646,7,0.41454023122787476,1.7869085289463746,5,0.5
-expression_strain,5677,5657,56,0.36287590861320496,2.107323304373448,5,0.5
-inducer,7889,7868,23,0.21219828724861145,2.264613330368638,5,0.5
-desalting_process,3720,3711,10,0.19111751019954681,2.611255039242421,5,0.5
-lysis_buffer,3714,3707,12,0.172078475356102,2.3357210421997974,5,0.5
-elution_buffer,3530,3525,21,0.054438717663288116,2.62212905402688,5,0.5
-plasmid,5480,5464,51,0.024756066501140594,3.375786331689511,5,0.5
-enzyme_name,7975,7911,234,-0.0832073986530304,3.3833475490469125,5,0.5
-medium_name,4806,4796,64,-0.09721468389034271,2.6786008337907408,5,0.5
-molecular_weight,295,294,1,,,5,0.5
-source_key,8021,8021,1,,,5,0.5
+organism_source,9834,9775,62,0.7354418635368347,1.6965245187920275,5,0.5
+induction_temperature,5383,5380,10,0.41430872678756714,1.9583334551521774,5,0.5
+expression_strain,8268,8257,51,0.31281837821006775,2.2561694054364247,5,0.5
+inducer,11954,11933,31,0.21848483383655548,2.57037572116691,5,0.5
+lysis_buffer,5625,5622,19,0.17804592847824097,2.309976794303761,5,0.5
+desalting_process,5540,5536,14,0.1778792291879654,2.726505566501225,5,0.5
+elution_buffer,5198,5186,25,0.0694195106625557,2.4611490534318428,5,0.5
+strain,3834,3783,63,0.0456826351583004,2.6767485257786485,5,0.5
+plasmid,8217,8202,64,0.021312322467565536,3.8060242263957456,5,0.5
+enzyme_name,12099,12026,306,-0.09378033876419067,3.4476950210926574,5,0.5
+medium_name,7644,7622,75,-0.09425070136785507,2.9421486267519406,5,0.5
+molecular_weight,495,491,1,,,5,0.5
+source_key,12179,12179,1,,,5,0.5
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_ALL_FIELDS.csv.gz
index 65dd754a461dfb68903d4a7400cc262ca2f8bb11..d5739ddc575155dacdaf153610f6b191ff5911dd 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ffc4863ee7ee6f430663df76f01100a752d874620ed9d580c9c6d23c121dfd9
-size 1686023
+oid sha256:7d7af43b4da1ed1dfb2702c8f706535fda84b8541c1103ebca3f5a5a414b2874
+size 2653748
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_FIELD_CLUSTER_METRICS.csv
index f4acd0a121b59633f1ca6697a110240ae2bb8c2d..d4222e4da794deb08bb8512535f93c51123e71d2 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.6_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,6117,68,0.8107371926307678,1.088422689947546,5,0.6
-strain,2444,2184,47,0.7107850313186646,1.4691608568152974,5,0.6
-inducer,7889,7841,36,0.7022331953048706,2.049736063241613,5,0.6
-expression_strain,5677,5574,60,0.5213732719421387,1.8855611195078856,5,0.6
-induction_temperature,3656,3645,19,0.20025323331356049,1.6497673452568422,5,0.6
-source_key,8021,8019,6,0.09909623861312866,3.8096047749079376,5,0.6
-plasmid,5480,5280,203,0.08477261662483215,2.7010186526133295,5,0.6
-molecular_weight,295,291,5,0.05576443672180176,2.405228213648723,5,0.6
-lysis_buffer,3714,3679,47,0.05248519033193588,2.299231758500828,5,0.6
-desalting_process,3720,3702,40,0.027227269485592842,2.7165719740592342,5,0.6
-enzyme_name,7975,7174,588,0.011261316016316414,2.6508562362810646,5,0.6
-elution_buffer,3530,3497,70,-0.04432287812232971,2.734217450381013,5,0.6
-medium_name,4806,4710,112,-0.068043053150177,2.6112269902503606,5,0.6
+organism_source,9834,9639,83,0.8320907950401306,1.229413022412698,5,0.6
+inducer,11954,11894,55,0.604241669178009,1.9748012538089221,5,0.6
+strain,3834,3544,94,0.40730932354927063,2.1512616511417773,5,0.6
+expression_strain,8268,8123,105,0.4014231860637665,2.1930546039241188,5,0.6
+induction_temperature,5383,5375,24,0.21695955097675323,1.723104855490049,5,0.6
+source_key,12179,12178,10,0.045199424028396606,3.655201084876773,5,0.6
+plasmid,8217,7959,341,0.03641128167510033,2.8032643605319456,5,0.6
+molecular_weight,495,483,6,0.023108134046196938,3.179936731640732,5,0.6
+lysis_buffer,5625,5579,69,0.011227880604565144,2.312488165867662,5,0.6
+desalting_process,5540,5495,56,0.003338330192491412,2.558065142656291,5,0.6
+enzyme_name,12099,11237,883,-0.007526148110628128,2.7476960364209204,5,0.6
+elution_buffer,5198,5158,90,-0.05552639067173004,2.5965222687973295,5,0.6
+medium_name,7644,7528,160,-0.08503243327140808,2.6905996197643027,5,0.6
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_ALL_FIELDS.csv.gz
index 31a6a661578f3f09bf51fad9faa5371d27b3ae08..8d89b89e634fc45678f310e219bab49a42064784 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f276c50157ce39ee486e6cf250fc0c5b91a53205287deda0f27c95ca8cd2892
-size 1689341
+oid sha256:cab3c8298419fab2b463d1a7638dea8be6dee4fdf2b8b77dfe4adc51b2f63b11
+size 2647240
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_FIELD_CLUSTER_METRICS.csv
index bc31eec6627d80a25256b85b0ecbeea02ff0ed11..c51d4db33adb2bedbe61a3cbd9ae553f0d5707e6 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.7_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5979,70,0.9046768546104431,0.6662478797358539,5,0.7
-strain,2444,2003,56,0.7933861017227173,1.027016593127256,5,0.7
-inducer,7889,7788,42,0.663221001625061,1.6345614764182945,5,0.7
-expression_strain,5677,5406,82,0.5707747936248779,1.0513776444862881,5,0.7
-enzyme_name,7975,5127,446,0.2601396143436432,1.8466009771077312,5,0.7
-plasmid,5480,4466,186,0.14653074741363525,2.07461954628352,5,0.7
-molecular_weight,295,263,17,0.06944355368614197,2.6093466988723426,5,0.7
-medium_name,4806,4425,169,0.04652847722172737,2.215995701687166,5,0.7
-lysis_buffer,3714,3563,110,-0.005272489972412586,2.3178370824185497,5,0.7
-induction_temperature,3656,3625,31,-0.008622316643595695,1.7727621670398925,5,0.7
-elution_buffer,3530,3342,125,-0.01700015924870968,2.553102043974565,5,0.7
-desalting_process,3720,3526,118,-0.0316702164709568,2.426816746186875,5,0.7
-source_key,8021,8006,109,-0.09172094613313675,2.3766081256362868,5,0.7
+organism_source,9834,9490,87,0.934995710849762,0.7748425686762738,5,0.7
+inducer,11954,11830,66,0.6362406611442566,1.847383782924625,5,0.7
+strain,3834,3105,97,0.5985273718833923,1.388472851140098,5,0.7
+expression_strain,8268,7782,149,0.5233995318412781,1.3594488507623532,5,0.7
+enzyme_name,12099,8105,735,0.2582234740257263,1.8682528067894666,5,0.7
+plasmid,8217,6561,324,0.14519894123077393,2.196227776943711,5,0.7
+molecular_weight,495,456,27,0.043932050466537476,2.676914533850017,5,0.7
+induction_temperature,5383,5340,41,0.01640150509774685,1.7534738503274196,5,0.7
+medium_name,7644,7172,263,-0.007269705645740032,2.2702773330934622,5,0.7
+elution_buffer,5198,4946,169,-0.03873402625322342,2.680737134817065,5,0.7
+lysis_buffer,5625,5454,166,-0.04855545237660408,2.426855043170666,5,0.7
+desalting_process,5540,5311,172,-0.05011814832687378,2.581706090890557,5,0.7
+source_key,12179,12168,144,-0.10621629655361176,2.5694946326938313,5,0.7
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_ALL_FIELDS.csv.gz
index 1ff9c3d9d6efab15304959d8f899ea29d37114df..dccc0fa3166800265238fa32afb6539ba49a710f 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52eaa19ace7b967c0c800cb335c1cf019becbd219c938e059374498c14740955
-size 1670051
+oid sha256:bf1d8cc3423ee5ab4d45735773c3bf26477b716484561d1a0b386d70e2532ad9
+size 2616448
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_FIELD_CLUSTER_METRICS.csv
index f8849cfee34898bdd258e6b75807c5c116e8b57c..c87908c59a2f192367f6ac4f72a3447b800a2035 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.8_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5861,75,0.9598073363304138,0.44336556918527825,5,0.8
-strain,2444,1855,50,0.9519201517105103,0.5636957403780866,5,0.8
-inducer,7889,7615,56,0.5611739754676819,1.7849241320741316,5,0.8
-expression_strain,5677,5233,80,0.5338608622550964,0.8995208371411649,5,0.8
-enzyme_name,7975,3081,322,0.4092719256877899,1.3713789658218687,5,0.8
-induction_temperature,3656,3526,42,0.40609872341156006,1.1228662543225003,5,0.8
-molecular_weight,295,178,17,0.29991239309310913,1.796784708345038,5,0.8
-medium_name,4806,3518,201,0.253959059715271,1.629324239525528,5,0.8
-plasmid,5480,3445,193,0.24020208418369293,1.488521839252724,5,0.8
-desalting_process,3720,2882,177,0.14139942824840546,2.0202743492527593,5,0.8
-elution_buffer,3530,2746,168,0.10630475729703903,1.957770439765363,5,0.8
-source_key,8021,7770,331,0.09236180782318115,1.975405320106651,5,0.8
-lysis_buffer,3714,2974,176,0.07679891586303711,2.022886315047004,5,0.8
+organism_source,9834,9327,91,0.9583161473274231,0.46456069446592557,5,0.8
+strain,3834,2782,73,0.7961609363555908,0.6357533527297731,5,0.8
+expression_strain,8268,7340,131,0.6295028924942017,0.9296579491933901,5,0.8
+inducer,11954,11611,107,0.5367256999015808,1.7571414370821283,5,0.8
+induction_temperature,5383,5235,57,0.5110968351364136,1.169708001254173,5,0.8
+enzyme_name,12099,4556,479,0.45910632610321045,1.273816986927687,5,0.8
+plasmid,8217,4735,281,0.25269994139671326,1.5412343878862145,5,0.8
+medium_name,7644,5760,330,0.23141148686408997,1.7789327516664015,5,0.8
+molecular_weight,495,347,34,0.18894857168197632,1.888895408862957,5,0.8
+desalting_process,5540,4271,260,0.10978957265615463,2.0963058721830907,5,0.8
+elution_buffer,5198,4030,246,0.0891888216137886,2.060596387071092,5,0.8
+lysis_buffer,5625,4633,268,0.05599077045917511,2.0999354627747873,5,0.8
+source_key,12179,11923,487,0.0509500578045845,2.059142146634667,5,0.8
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_ALL_FIELDS.csv.gz b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_ALL_FIELDS.csv.gz
index 6e272f955f55c017f894ae0ad16c228d3dba5ad6..a1ebeedaa27f7112a603db74996adabcf9ec78b7 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_ALL_FIELDS.csv.gz
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_ALL_FIELDS.csv.gz
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c7394d3be1200a8bec79c4de16506329f22190b0d5346217cfdcdaec34aa535
-size 1568355
+oid sha256:a5059c48537a455eb273fc37ab24cd284cf6ab059326eb7e5b27b5a7753e150b
+size 2430253
diff --git a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_FIELD_CLUSTER_METRICS.csv b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_FIELD_CLUSTER_METRICS.csv
index 301567fc819e5b9572005fe92be3d23155f0c1f8..f52e36c09d1dd1ef68563e63c6979592d73aa52f 100644
--- a/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_FIELD_CLUSTER_METRICS.csv
+++ b/data/clustering/model=cambridgeltl__SapBERT-from-PubMedBERT-fulltext/min=5/t=0.9_FIELD_CLUSTER_METRICS.csv
@@ -1,14 +1,14 @@
 field,n_total,n_used,n_clusters,silhouette_cosine,davies_bouldin,min_community_size,threshold
-organism_source,6292,5805,76,0.9988065958023071,0.21391314547085466,5,0.9
-strain,2444,1774,45,0.9857630729675293,0.25944007703589106,5,0.9
-induction_temperature,3656,3376,55,0.9228525161743164,0.40443169118805516,5,0.9
-source_key,8021,4724,569,0.8441194295883179,0.6007825077179415,5,0.9
-molecular_weight,295,83,11,0.7941944003105164,0.6574338997581686,5,0.9
-enzyme_name,7975,1192,153,0.7630952000617981,0.7228691971900033,5,0.9
-desalting_process,3720,1287,144,0.7520548105239868,0.7430092355083585,5,0.9
-inducer,7889,7345,152,0.6572569012641907,1.5904128266536601,5,0.9
-medium_name,4806,2372,175,0.6524311900138855,0.6996787893911582,5,0.9
-expression_strain,5677,5055,118,0.6522308588027954,0.6353441522018386,5,0.9
-lysis_buffer,3714,1502,170,0.6470782160758972,0.9098308477788016,5,0.9
-plasmid,5480,2489,151,0.6417360305786133,0.6889688012440343,5,0.9
-elution_buffer,3530,1537,146,0.5825046300888062,0.9490314519584871,5,0.9
+organism_source,9834,9248,92,0.9984946250915527,0.2221954156206894,5,0.9
+strain,3834,2646,68,0.9323930740356445,0.34673557053547144,5,0.9
+induction_temperature,5383,4998,61,0.9313704967498779,0.44591222788448137,5,0.9
+enzyme_name,12099,1791,221,0.7934830784797668,0.666998703828258,5,0.9
+molecular_weight,495,181,24,0.7922830581665039,0.6710872325997466,5,0.9
+source_key,12179,7341,853,0.7631043195724487,0.7203005314710007,5,0.9
+desalting_process,5540,1802,197,0.7026733160018921,0.833936524248297,5,0.9
+expression_strain,8268,6978,162,0.6935133934020996,0.5830769840042045,5,0.9
+inducer,11954,11103,206,0.6429955363273621,1.6188702786688849,5,0.9
+plasmid,8217,3184,199,0.6202341914176941,0.6993883152306856,5,0.9
+medium_name,7644,3826,255,0.6149197220802307,0.7474718763619609,5,0.9
+lysis_buffer,5625,2311,253,0.6131700277328491,0.9329422684838516,5,0.9
+elution_buffer,5198,2069,192,0.5402795672416687,1.0334736451137767,5,0.9