Integrate finetuning & fingerprinting

README.md

@@ -35,8 +35,6 @@ Visit https://graphium-docs.datamol.io/.
 
 ## Installation for developers
 
-### For CPU and GPU developers
-
 Use [`mamba`](https://github.com/mamba-org/mamba), a faster and better alternative to `conda`.
 
 If you are using a GPU, we recommend enforcing the CUDA version that you need with `CONDA_OVERRIDE_CUDA=XX.X`.
@@ -53,18 +51,6 @@ mamba activate graphium
 pip install --no-deps -e .
 ```
 
-### For IPU developers
-```bash
-# Install Graphcore's SDK and Graphium dependencies in a new environment called `.graphium_ipu`
-./install_ipu.sh .graphium_ipu
-```
-
-The above step needs to be done once. After that, enable the SDK and the environment as follows:
-
-```bash
-source enable_ipu.sh .graphium_ipu
-```
-
 ## Training a model
 
 To learn how to train a model, we invite you to look at the documentation, or the jupyter notebooks available [here](https://github.com/datamol-io/graphium/tree/master/docs/tutorials/model_training).
@@ -148,6 +134,39 @@ graphium-train --config-path [PATH] --config-name [CONFIG]
 ```
 Thanks to the modular nature of `hydra` you can reuse many of our config settings for your own experiments with Graphium.
 
+### Finetuning
+
+After pretraining a model and saving a model checkpoint, the model can be finetuned to a new task
+
+```bash
+graphium-train +finetuning [example-custom OR example-tdc] finetuning.pretrained_model=[model_identifier]
+```
+
+The `[model_identifier]` serves to identify the pretrained model among those maintained in the `GRAPHIUM_PRETRAINED_MODELS_DICT` in `graphium/utils/spaces.py`, where the `[model_identifier]` maps to the location of the checkpoint of the pretrained model.
+
+We have provided two example yaml configs under `expts/hydra-configs/finetuning` for finetuning on a custom dataset (`example-custom.yaml`) or for a task from the TDC benchmark collection (`example-tdc.yaml`).
+
+When using `example-custom.yaml`, to finetune on a custom dataset, we nee to provide the location of the data (`constants.data_path=[path_to_data]`) and the type of task (`constants.task_type=[cls OR reg]`).
+
+When using `example-tdc.yaml`, to finetune on a TDC task, we only need to provide the task name (`constants.task=[task_name]`) and the task type is inferred automatically.
+
+Custom datasets to finetune from consist of two files `raw.csv` and `split.csv`. The `raw.csv` contains two columns, namely `smiles` with the smiles strings, and `target` with the corresponding targets. In `split.csv`, three columns `train`, `val`, `test` contain the indices of the rows in `raw.csv`. Examples can be found under `expts/data/finetuning_example-reg` (regression) and `expts/data/finetuning_example-cls` (binary classification).
+
+### Fingerprinting
+
+Alternatively, we can also obtain molecular embeddings (fingerprints) from a pretrained model:
+```bash
+graphium fps create [example-custom OR example-tdc] pretrained.model=[model_identifier] pretrained.layers=[layer_identifiers]
+```
+
+We have provided two example yaml configs under `expts/hydra-configs/fingerprinting` for extracting fingerprints for a custom dataset (`example-custom.yaml`) or for a dataset from the TDC benchmark collection (`expample-tdc.yaml`).
+
+After specifiying the `[model_identifier]`, we need to provide a list of layers from that model where we want to read out embeddings via `[layer_identifiers]` (which requires knowledge of the architecture of the pretrained model).
+
+When using `example-custom.yaml`, the location of the smiles to be embedded needs to be passed via `datamodule.df_path=[path_to_data]`. The data can be passed as a csv/parquet file with a column `smiles`, similar to `expts/data/finetuning_example-reg/raw.csv`.
+
+When extracting fingerprints for a TDC task using `expample-tdc.yaml`, we need to specify `datamodule.benchmark` and `datamodule.task` instead of `datamodule.df_path`.
+
 ## License
 
 Under the Apache-2.0 license. See [LICENSE](LICENSE).

docs/api/graphium.finetuning.md

@@ -10,4 +10,4 @@ Module for finetuning models and doing linear probing (fingerprinting).
 
 ::: graphium.finetuning.finetuning_architecture.FinetuningHead
 
-::: graphium.finetuning.fingerprinting.Fingerprinter
+::: graphium.fingerprinting.fingerprinter.Fingerprinter

docs/cli/graphium-train.md

@@ -24,7 +24,7 @@ graphium-train architecture=toymix tasks=toymix training=toymix model=gcn accele
 automatically selects the correct configs to run the experiment on GPU.
 Finally, you can also run a fine-tuning loop:
 ```bash
-graphium-train +finetuning=admet
+graphium-train +finetuning=example-tdc
 ```
 
 To use a config file you built from scratch you can run

docs/cli/graphium.md

@@ -103,7 +103,7 @@ $ graphium finetune [OPTIONS] COMMAND [ARGS]...
 
 **Commands**:
 
-* `admet`: Utility CLI to easily fine-tune a model on...
+* `tdc`: Utility CLI to easily fine-tune a model on...
 * `fingerprint`: Endpoint for getting fingerprints from a...
 
 ### `graphium finetune admet`
@@ -135,7 +135,7 @@ Endpoint for getting fingerprints from a pretrained model.
 The pretrained model should be a `.ckpt` path or pre-specified, named model within Graphium.
 The fingerprint layer specification should be of the format `module:layer`.
 If specified as a list, the fingerprints from all the specified layers will be concatenated.
-See the docs of the `graphium.finetuning.fingerprinting.Fingerprinter` class for more info.
+See the docs of the `graphium.fingerprinting.fingerprinter.Fingerprinter` class for more info.
 
 **Usage**:
 

expts/data/finetuning_example-cls/raw.csv

@@ -0,0 +1,201 @@
+,Drug_ID,smiles,target
+0,644675,CC(=O)N(c1ccc2oc(=O)sc2c1)S(=O)(=O)c1cccs1,0
+1,644890,COc1ccccc1C(c1nnnn1C(C)(C)C)N1CCN(Cc2ccncc2)CC1,1
+2,645164,CCC(c1nnnn1CC1CCCO1)N(CCN1CCOCC1)Cc1cc2cc(C)ccc2[nH]c1=O,0
+3,6602688,Br.N=c1n(CCN2CCOCC2)c2ccccc2n1CC(=O)c1ccc(Cl)c(Cl)c1,1
+4,645448,CCC(C)(C)NC(=O)c1ccc2c(c1)N(CC(=O)OC)C(=O)C(C)(C)O2,0
+5,645569,CCc1cc2c(=O)[nH]cnc2s1,0
+6,645818,COc1cccc2c(=O)c(C(=O)NCc3cccs3)c[nH]c12,1
+7,645911,CCc1nnc(SCc2ccc(OC(C)C)cc2)n1N,0
+8,645965,Cc1ccc2c(c1)nnn2C1CCN(CC(=O)N2c3ccccc3CC2C)CC1,1
+9,646164,CCOC(=O)CSC1=C(C#N)C(C)C2=C(CCCC2=O)N1,0
+10,646293,Cc1ccc2cc(C)c3nnc(SCC(=O)NCc4ccco4)n3c2c1,0
+11,646353,CCOC(=O)N1CCN(S(=O)(=O)Cc2ccccc2)CC1,0
+12,646472,CCOC(=O)c1cc2sc(C)cc2n1CC(=O)N1CCN(C(=O)c2ccco2)CC1,0
+13,646515,CCOC(=O)C(NC(C)=O)C(OC(C)=O)c1cccc(N(CCO)CCO)c1,0
+14,646597,CCC(=O)Nc1cc2c(cc1C(=O)c1ccccc1)OCCO2,0
+15,6602690,CN(C)CCNC(=O)C(C(=O)c1ccc(F)cc1)n1ccccc1=O.Cl,0
+16,646768,Cc1cc(C)nc(N2CCC(C(=O)NCCc3ccc(F)cc3)CC2)n1,0
+17,646780,COc1cc(-c2nnc(-c3ccc(N4CCOCC4)cc3)o2)cc(OC)c1OC,0
+18,646897,CN(C)C=C1C(=O)N(C2CCCCC2)C(=O)N(C2CCCCC2)C1=O,0
+19,646955,COC(=O)CN(c1ccccn1)S(=O)(=O)c1ccccc1,0
+20,6398903,Cc1ccc(C)c(/C(O)=C2/C(=O)C(=O)N(CCN3CCOCC3)C2c2ccco2)c1,0
+21,647114,Cn1c(SC2=CS(=O)(=O)c3ccccc32)nc2ccccc21,0
+22,647205,CC(C(=O)NC1CCCC1)N(C(=O)c1snc(C(N)=O)c1N)c1ccc2c(c1)OCCO2,1
+23,647430,Cc1cc(C)n(-c2nc3c(c(=O)[nH]c(=O)n3C)n2C(C)C)n1,0
+24,647727,O=C(Nc1ccc(S(=O)(=O)N2CCCC2)cc1)c1ccc(CN2CCOCC2)cc1,1
+25,6602522,Cl.O=C(CN1CCN(c2ncccn2)CC1)NCCC1=CCCCC1,1
+26,647937,CC(=O)NC1(c2cccc(F)c2)CCN(CC(=O)NC2CCCCC2)CC1,0
+27,647996,Cc1cccc(Nc2nc3c(c(=O)n(C)c(=O)n3C)n2CC(O)CO)c1,0
+28,648175,CCn1c(Cc2ccccc2)nnc1SCC(=O)NC(C)(C)C,0
+29,648282,COc1cc2cc(CN(CCCO)S(=O)(=O)c3ccccc3Cl)c(=O)[nH]c2cc1OC,0
+30,648407,CCc1nnc(NC(=O)CSc2nnc(COc3ccccc3)n2Cc2ccccc2)s1,0
+31,648481,CSc1nc2nc3c(c(=O)n2[nH]1)CN(Cc1ccccc1)CC3,0
+32,648708,Cc1nn(C)c(C)c1CNC(=O)c1cnn2c1NC(c1ccccc1)CC2C(F)(F)F,0
+33,648836,Cc1ccc(C(c2nnnn2CC2CCCO2)N2CCN(C(=O)c3ccco3)CC2)cc1,0
+34,648878,CCN(CC)c1ccc2c(Cl)c(Br)c(=O)oc2c1,0
+35,648947,COc1ccc(OC)c(NC(=O)C(CC(=O)O)NCc2ccco2)c1,0
+36,649015,O=C1CC(c2cccs2)c2cc3c(cc2N1)OCO3,1
+37,649453,CCn1c(COc2ccccc2)nnc1SCC(=O)Nc1cc(OC)ccc1OC,0
+38,649754,Nc1ncc(-c2ccccc2)n1CC1CCCO1,1
+39,649786,CCCCn1c(SCC(=O)N2CCCC2)nc2c1c(=O)n(C)c(=O)n2C,0
+40,649878,O=C(O)CCn1nnc(-c2cccs2)n1,0
+41,650002,Cn1c(=O)[nH]c(=O)c2c1nc(NCCCO)n2CCCc1ccccc1,0
+42,650100,COc1ccccc1OCCn1cc(C(=O)c2ccco2)c2ccccc21,0
+43,650250,Cc1ccc(-c2csc(N3CCC(NS(=O)(=O)c4ccc5c(c4)OCCO5)CC3)n2)cc1,0
+44,650341,COc1ccc(S(=O)(=O)N2CCC(NC(=O)Nc3ccc(C)cc3)CC2)cc1,0
+45,650486,Cn1c(CNC(=O)Nc2ccccc2)nnc1SCc1ccccc1,0
+46,650558,O=C(CSc1nnc(CNc2ccccc2)o1)N1CCCc2ccccc21,0
+47,650691,CCCCNC(=O)NS(=O)(=O)c1ccc(C(=O)OC(C)C)o1,0
+48,6602999,CCOC(=O)C1Cc2c([nH]c3ccccc23)CN1.Cl,1
+49,650985,COc1ccc(CCNc2c(C)c(C)nc3ncnn23)cc1,1
+50,5768893,COCCN1C(=O)C(=O)/C(=C(/O)c2cccc(OC)c2)C1c1ccco1,0
+51,651076,CCC(=O)Nc1cccc(NC(=O)CSc2nnnn2Cc2ccccc2)c1,0
+52,651205,CCc1c(C)nc2c(C#N)c(C)[nH]n2c1=O,0
+53,651338,CCC(=O)N(Cc1ccco1)c1nc(-c2ccccc2)cs1,0
+54,651587,O=C(OCCCN1CCCCC1)c1ccc(O)cc1,1
+55,651589,Cc1cc(NC(=O)CCC(=O)N2CCC3(CC2)OCCO3)no1,0
+56,651769,O=C(CNC(=O)c1ccco1)OCc1c(F)cccc1Cl,0
+57,652002,O=C(CSc1nnc(-c2cccnc2)o1)Oc1ccccc1,0
+58,652521,CN(C)S(=O)(=O)c1ccc(C(=O)Nc2ccc(CN3CCCC3)cc2)cc1,1
+59,652549,CC(C)C(C(=O)NCC1CCCO1)N(Cc1ccco1)C(=O)CNS(=O)(=O)c1ccc(F)cc1,0
+60,652700,COCCNC(=O)COC(=O)c1nsc(Cl)c1Cl,0
+61,652799,CCOc1ccc(-c2nnn(CC(=O)Nc3cc(OC)ccc3OC)n2)cc1OCC,0
+62,6603138,Cl.O=C(CN1CCN(C2CCCCC2)CC1)NCCC1=CCCCC1,1
+63,653279,O=C(O)C1C2C=CC3(CN(Cc4ccccn4)C(=O)C13)O2,0
+64,653412,Cn1c(SCC(=O)NCc2ccc3c(c2)OCO3)nnc1-c1cc2ccccc2cc1O,1
+65,6603457,CCC(c1nnnn1CCOC)N1CCN(C(=O)c2ccco2)CC1.Cl,0
+66,653646,O=S(=O)(c1ccccc1)N1CCN(c2cc(-c3ccccc3)nc3ncnn23)CC1,0
+67,653695,O=C(CSc1n[nH]c(-c2ccccc2O)n1)N1CCCc2ccccc21,1
+68,653778,COc1ccccc1-n1c(SCC(=O)N2CCCC2)nc2cccnc21,0
+69,653799,CCc1nnc2sc(-c3ccc(NC(=O)c4ccco4)cc3)nn12,0
+70,653914,COc1ccc(-c2nnn(CC(=O)N(CC(=O)NCCC(C)C)Cc3cccs3)n2)cc1OC,1
+71,654078,CCn1c(SCc2ccc(C#N)cc2)nnc1-c1ccc(S(=O)(=O)N2CCCCC2)cc1,0
+72,654182,CCOC(=O)Cc1cc(=O)n2[nH]c(C)c(-c3ccccc3)c2n1,0
+73,6398932,CCOc1cccc(/C(O)=C2/C(=O)C(=O)N(Cc3ccco3)C2c2ccncc2)c1,0
+74,654363,O=C(NC1CCCCC1)C(c1cccs1)N(Cc1cccs1)C(=O)c1ccco1,1
+75,654435,O=C(CSc1nc2ccccc2o1)Nc1nc2ccccc2s1,0
+76,5373216,COc1n[nH]c2nncnc12,0
+77,654546,CCc1ccc(N2CC(C(=O)NC3=NCCS3)CC2=O)cc1,0
+78,654623,COc1ccc(CNC(=O)CN(CC2CCCO2)C(=O)CNS(=O)(=O)c2ccccc2)cc1,0
+79,654635,COc1cc(C2C(C(=O)c3ccc(C)o3)=C(O)C(=O)N2CCc2ccccc2)ccc1O,0
+80,654761,Cc1ccc(C)n1C(Cc1ccccc1)C(=O)O,0
+81,655183,COc1ccccc1CN(Cc1cc2cc(C)cc(C)c2[nH]c1=O)Cc1nnnn1CC1CCCO1,0
+82,655265,COc1ccc(OCc2nnc(SCC(=O)O)n2N)cc1,0
+83,5768421,COCCCN1C(=O)C(=O)/C(=C(/O)c2ccc(OC(C)C)c(C)c2)C1c1ccncc1,0
+84,655401,CCN(C1CCCCC1)S(=O)(=O)c1ccc(S(=O)(=O)NCc2ccncc2)cc1,1
+85,655439,CCOC(=O)C1=C(C)NC(=O)NC1c1ccoc1,0
+86,655857,c1ccc(Cn2nnc3c(N4CCc5ccccc5C4)ncnc32)cc1,0
+87,655866,COc1ccc(C(=O)NC2CC3CCCC(C2)N3CC(C)C)cc1OC,1
+88,655948,Cc1cc(N2CCN(c3nc4ccccc4s3)CC2)n2ncnc2n1,0
+89,656017,CC(C)C(=O)Nc1cc2c(cc1C(=O)c1ccccc1)OCCO2,0
+90,656027,COc1ccc(CCN2C(=O)C(O)=C(C(=O)c3ccco3)C2c2cccs2)cc1OC,0
+91,656095,COc1ccc(C2C(C(=O)N3CCOCC3)=C(C)NC3=C2C(=O)CC(C)(C)C3)c(OC)c1,0
+92,656157,O=c1oc(-c2ccco2)nc2c1cnn2-c1ccccc1,0
+93,656183,CC(C)OC(=O)NCCOC(=O)Nc1cccc(Cl)c1,0
+94,656257,Cc1ccccc1OCC1Cn2c(nc3c2c(=O)[nH]c(=O)n3C)O1,0
+95,656272,Cc1ccc(-c2[nH]n3c(=O)c4c(nc3c2C)CCCC4)cc1,1
+96,656290,Cc1cccc(NC(=O)Cn2c(=O)oc3ccccc32)c1,1
+97,6603060,I.OCCNC1=NCCN1,0
+98,6449251,COc1cc2c(cc1OC)/C(=C/C(=O)N1CCOCC1)NC(C)(C)C2.Cl,1
+99,135449532,Cc1cc(=O)[nH]c(-n2nc(C)cc2C)n1,0
+100,5940036,CCOC(=O)C1=C(N)n2c(s/c(=C\c3ccco3)c2=O)=C(C(=O)OCC)C1c1ccco1,0
+101,208296,O=c1nc(-c2ccccc2)cn[nH]1,0
+102,658411,COC(=O)c1ccc(Oc2cc(C)nc(-n3nc(C)cc3C)n2)cc1,0
+103,658723,COC(=O)c1[nH]c2ccc(Br)cc2c1NC(=O)CN1CCN(C2CCCCC2)CC1,1
+104,658813,CCc1c(C)c(C#N)c2nc3ccccc3n2c1Nc1c(C)n(C)n(-c2ccccc2)c1=O,0
+105,135415833,O=c1c(Cc2ccccc2)c(O)nc2n1CCS2,0
+106,658879,CCC(NC(=O)Nc1cc(OC)c(OC)c(OC)c1)(C(F)(F)F)C(F)(F)F,0
+107,659040,O=c1[nH]c(=S)[nH]nc1Cc1ccccc1,0
+108,135435901,COc1ccc(C2CC(=O)C(C=NCCN3CCOCC3)=C(O)C2)cc1,0
+109,16411130,Cc1nc(/N=C(\N)Nc2ccccc2)nc2ccccc12,0
+110,659321,CCOc1ccc(CSC(CC(=O)O)C(=O)O)cc1,0
+111,2838016,CC1=CC=CN2CC(O)CN=C12.Cl,0
+112,6603569,Cc1ccc2c(c1)[C@@H]1CN(C)CC[C@@H]1N2S(=O)(=O)c1ccc(F)cc1.Cl,1
+113,1922089,Cc1cc(C)c(-n2c(O)c(C=NCCN3CCOCC3)c(=O)[nH]c2=O)c(C)c1,0
+114,659756,O=C1CCCN1CC(CN1CCOCC1)Sc1nnnn1-c1ccccc1,0
+115,660120,Cc1cc(-c2cc(-c3ccc(Cl)cc3)nc(N)c2C#N)co1,0
+116,660285,COc1ccc(S(=O)(=O)N2CCC(N3CCCCC3)CC2)cc1,1
+117,660304,O=C(COc1ccccc1)Nc1ccc(-c2nnc(-c3ccco3)o2)cc1,0
+118,5389248,COc1ccc(C2/C(=C(/O)c3ccc(Cl)cc3)C(=O)C(=O)N2CCN2CCOCC2)c(OC)c1,0
+119,5389254,Cc1oc2cc(O)ccc2c(=O)c1-c1cnn(-c2ccccc2)c1,0
+120,660546,Cc1cccc(OCC(=O)N2CCC(N3CCCCCC3)CC2)c1,1
+121,135420605,CCCCc1c(O)nc(SCCN(C)C)n(-c2ccccc2)c1=O,0
+122,660831,COC1(OC)N=C(NC(=O)Nc2ccccc2)C2(C#N)C(c3ccccc3)C12C#N,0
+123,660995,Cc1ccc(S(=O)(=O)NCCSc2nnnn2C)cc1,0
+124,661065,CNc1oc(-c2cccs2)nc1C#N,0
+125,661098,CC(=O)Nc1ccc(Nc2ncnc3c2cnn3-c2ccccc2)cc1,0
+126,661170,O=C(Cc1ccc(Cl)cc1)Nc1cccc(-c2nnc(-c3ccco3)o2)c1,0
+127,661178,CCCOc1ccc(CSC(CC(=O)O)C(=O)O)cc1,0
+128,661187,c1ccc(-n2ncc3c(NCCCN4CCOCC4)ncnc32)cc1,0
+129,661203,CC1CCc2cccc3c2N1c1cc(C#N)c(C#N)cc1O3,0
+130,661217,N#Cc1nc(-c2cccs2)oc1NCc1ccccc1,1
+131,661296,CCc1c(C(=O)O)[nH]c2ccc(Br)cc12,0
+132,661300,S=c1nc(-c2ccccc2)[nH]n1-c1ccccc1,0
+133,661349,CC(=O)c1c(C(C)=O)c(C)n(NC(=O)c2ccncc2)c1C,0
+134,661355,Cn1c(=O)n(CCC(=O)O)c2ccccc21,0
+135,661406,CC1(C)CC(=O)C(CCCN2C(=O)c3ccccc3C2=O)C(=O)C1,0
+136,6603015,CN(C)CC(O)COc1cccc(OCC(O)CN(C)C)c1.Cl,0
+137,6603014,CN(C)CC(O)COc1ccc(C(C)(C)c2ccc(OCC(O)CN(C)C)cc2)cc1.Cl,0
+138,661455,Oc1ccccc1CNn1cnnc1,0
+139,661513,CCn1c(SCC(=O)Nc2ccccc2C(=O)OC)nnc1-c1ccc(N)cc1,0
+140,661518,CCc1cc2c(=O)c(-c3nc4ccccc4[nH]3)coc2cc1OC(=O)N1CCOCC1,0
+141,661528,Cc1ccc(C2Nc3ccccc3C(=O)N2Cc2ccco2)cc1,0
+142,661552,Cc1cn2c(-c3ccncc3)nnc2s1,0
+143,661761,CCOC(=O)c1[nH]c2cc3c(cc2c1NC(=O)CN1CCc2ccccc2C1)OCO3,1
+144,5389368,COc1ccc(/C(O)=C2\C(=O)C(=O)N(c3cc(C)on3)C2c2ccc(OC)c(OC)c2)cc1OC,0
+145,5389389,COc1cccc(/C(O)=C2/C(=O)C(=O)N(c3cc(C)on3)C2c2cccs2)c1,0
+146,5389423,CCN(CC)CCN1C(=O)C(=O)/C(=C(/O)c2cccc(OC)c2)C1c1ccccn1,0
+147,661999,COc1cccc(C2C(C(=O)c3cc4ccccc4o3)=C(O)C(=O)N2c2cc(C)on2)c1OC,0
+148,662011,CCN1CCCC1Cn1cnc2c([nH]c3ccc(C)cc32)c1=O,1
+149,6881185,COCCCNC(=O)c1c(N)n(/N=C/c2ccccn2)c2nc3ccccc3nc12,0
+150,6881246,CCO/C(C)=N/n1c2nc3ccccc3nc2c2c(=O)n(CC(C)C)c(C)nc21,0
+151,662144,CCOC(=O)c1c(C)n(C)c2ccc(OC)c(NC(=O)CN3CCN(Cc4ccccc4)CC3)c12,1
+152,662340,CCOC(=O)c1cc2c(=O)n3cccc(C)c3nc2n(CCCOC)c1=NC(=O)c1cccnc1,0
+153,5389504,CCOc1ccc(/C(O)=C2/C(=O)C(=O)N(CCOCCO)C2c2cccnc2)cc1,0
+154,662407,NC(=O)C1CCN(C(=O)CN2C(=O)c3ccccc3S2(=O)=O)CC1,0
+155,662515,CCOC(=O)c1[nH]c2cc(OC)c(OC)cc2c1NC(=O)c1nonc1C,0
+156,6603499,COc1ccc(C(=O)OC(C)CN2CCN(C)CC2)cc1OC.Cl,1
+157,9614332,C[n+]1cccc(CNC(=O)/C=N/O)c1.[I-],0
+158,662647,COc1ccccc1-c1nnc2n1N=C(c1ccc(O)c(O)c1)CS2,0
+159,662710,COc1cccc(-c2nnc3sc(-c4ccc(C)cc4)nn23)c1,0
+160,662745,CCOCCCn1cnc2c([nH]c3cc(OC)ccc32)c1=O,0
+161,200556,Cl.OC1(c2ccc(F)cc2)CCNC1,0
+162,662794,CC(C)(C)OC(=O)NCCc1nnc(SCC(=O)Nc2cccc(Cl)c2)o1,0
+163,662799,O=C1C2ON(c3ccccc3)C(c3ccncc3)C2C(=O)N1c1ccccc1,0
+164,662838,CCCn1nc(NC(=O)CC(C)C)c2cc3ccccc3nc21,0
+165,662878,CCOc1ccccc1NC(=O)CSc1nnc(-c2cnccn2)n1C,1
+166,662996,CCOc1ccc(-c2nnc3n2N=C(c2ccc(OC)cc2)CS3)cc1,0
+167,663008,CC(C)COP(=O)(c1ccc(N(C)C)cc1)C(O)c1ccccc1F,0
+168,6603621,Cl.NCCCCCc1nnc(SCc2ccccc2Cl)o1,1
+169,663121,Nc1c(S(=O)(=O)c2ccccc2)c2nc3ccccc3nc2n1Cc1ccco1,1
+170,9615342,Cn1c[n+](C)cc1/C=N/O.[I-],0
+171,663125,Oc1ccc(-c2[nH]ncc2-c2ccc(Cl)cc2)c(O)c1,1
+172,663143,CCOc1ccc(C2=Nn3c(nnc3-c3ccccc3OC)SC2)cc1,0
+173,663146,COc1ccc2c(c1)[nH]c1c(N3CCN(Cc4ccc5c(c4)OCO5)CC3)ncnc12,1
+174,663168,COc1cccc(-c2nnc3n2N=C(C(C)(C)C)CS3)c1,0
+175,663337,CCOC(=O)c1[nH]c2cc(OC)c(OC)cc2c1NC(=O)c1ccc2c(c1)OCO2,0
+176,663340,COc1ccc(CCn2c(=N)c(C(=O)NCc3ccco3)cc3c(=O)n4ccccc4nc32)cc1,1
+177,663539,Cc1nc(SCC(=O)Nc2ccc3c(c2)OCCO3)c2oc3ccccc3c2n1,0
+178,663581,COC(=O)[C@@H](NC(=O)Nc1ccc(C(C)=O)cc1)C(C)C,0
+179,5389740,CCCN1C(=O)C2(/C(=C(\O)c3ccc4c(c3)OCCO4)C(=O)C(=O)N2CCCOC)c2ccccc21,0
+180,663736,COC(=O)[C@H](Cc1ccccc1)NC(=O)N1CCN(Cc2ccccc2)CC1,1
+181,663792,CCOCCCn1c(=N)c(C(=O)NCc2ccc3c(c2)OCO3)cc2c(=O)n3cccc(C)c3nc21,1
+182,54676164,CCOC(=O)C1=C(O)C(=O)N(c2ccc(S(N)(=O)=O)cc2)C1c1ccc(OC)cc1,0
+183,664033,CC1(C)CCCN(C(=O)c2coc(=O)c(Br)c2)C1,0
+184,664154,COC(=O)c1[nH]c2cc(C)ccc2c1NC(=O)CN1CCCc2ccccc21,1
+185,5389802,Cc1nc2ccccn2c1/C(O)=C1\C(=O)C(=O)N(CCCn2ccnc2)C1c1ccncc1,0
+186,664250,CC(C)(C)OC(=O)N1CCCC1C(=O)NCCc1ccccc1,0
+187,135513628,CCOC(=O)/C(C(N)=NCCCO)=C(\O)OCC,0
+188,664461,O=C1COc2ccc(OCc3ccc(F)cc3)cc21,0
+189,6603365,Cl.c1ccc2c(c1)oc1c(NCCCn3ccnc3)ncnc12,1
+190,5389869,COc1ccc(-c2c(C)oc3c(CN4CCN(CCO)CC4)c(O)ccc3c2=O)cc1OC,0
+191,5389875,CC1Cc2cc(/C(O)=C3/C(=O)C(=O)N(CCN4CCOCC4)C3c3cccc(Cl)c3)ccc2O1,0
+192,5389878,COc1ccc(C2/C(=C(/O)c3ccc4c(c3)CC(C)O4)C(=O)C(=O)N2CCN2CCOCC2)cc1,0
+193,664733,Cc1nc2c3cnn(-c4ccc(C)c(C)c4)c3ncn2n1,0
+194,664737,CCn1c(=N)c(S(=O)(=O)c2ccc(F)cc2)cc2c(=O)n3ccccc3nc21,1
+195,664759,O=C(CN1CCN(Cc2ccccc2)CC1)c1ccc(Br)cc1,1
+196,5389891,C/C(Cl)=C\Cn1c(N2CCCC2)nc2c1c(=O)[nH]c(=O)n2C,0
+197,664983,CCc1ccc(N2CC(C)Cn3c2nc2c3c(=O)n(CCN3CCOCC3)c(=O)n2C)cc1,0
+198,6603255,Br.CC1(C(=O)CSc2nc3ccccc3s2)CCC(=O)O1,0
+199,665081,Cn1c(-c2ccc(CN3CCCCC3)o2)nc2ccccc21,0