Sponsored by
PoYo.AI
genbio-ai / GB.Tissue-3M
Model's Last Updated: July 26 2026
Introduction of GB.Tissue-3M
Model Details of GB.Tissue-3M
GB.Tissue-3M
For more details, please refer to https://huggingface.co/genbio-ai/GB.Tissue-60M .
Finetuning GB.Tissue for spatial single cell downstream tasks
We introduce how to finetune and evaluate our pre-trained GB.Tissue foundation models for downstream tasks. These tasks can be classified into the following categories:
- Cell-level classification tasks : niche label type prediction
- Cell-level regression tasks : cell density prediction
Note: All the following scripts should be run under
ModelGenerator/
.
Download data
The related data is deposited at
https://huggingface.co/datasets/genbio-ai/tissue-downstream-tasks
. Please download the data and put under
ModelGenerator/downloads
as
cell_density
or
niche_type_classification
. Under each sub-directory, there are three files denote different split (xx.train.h5ad, xx.val.h5ad, xx.test.h5ad).
For each
.h5ad
, several obs attributes should be included to reprezent the spatial (coordinate) information (like
x
,
y
), the label information (like
niche_label
). All the column fields will be specified in the following
config.yaml
file.
Note: the file
scRNA_genename_and_index.tsv
includes all the corresponding gene name and index in h5ad file.
Cell-level classification tasks
niche label type prediction
We fully finetune GB.Tissue for niche label type prediction.
Finetuning script
CUDA_VISIBLE_DEVICES=7 nohup mgen fit --config experiments/GB.Tissue/niche_type_classfification.yaml > logs/nohup/GB.Tissue.niche_type_classfification.yaml.log 2>&1 &
Note:
The
filter_columns
includes label column and spatial coordinate column.
rename_columns
keep unchanged and will be used for running.
Evaluation script
Once finished run, there will be several
ckpt
file under the specified output directory
default_root_dir
. Then we can use the
ckpt
to evaluate on test dataset.
CUDA_VISIBLE_DEVICES=6 nohup mgen test --config experiments/GB.Tissue/niche_type_classfification.yaml \
--ckpt_path ckpt_path \
> ckpt_path.pred.log 2>&1 &
Note:
ckpt_path
is the finetuned checkpoint path.
Cell-level regression tasks
cell density prediction
The config file is like
experiments/GB.Tissue/cell_density_regression.yaml
, all the fintuning running and evaluation are similar as classification task.
Dump embedding
We can dump embedding for a
.h5ad
file. The script is as:
CUDA_VISIBLE_DEVICES=3 nohup mgen predict --config experiments/GB.Tissue/emb.xenium.yaml > logs/nohup/GB.Tissue.emb.xenium.log 2>&1 &
The output file will be under specified
output_dir
like
./logs/emb.xenium/lightning_logs/pred_output
. Each batch will be saved and a merged one will also be generated as
predict_predictions.pt
. The
predict_predictions.pt
file satcks all batches:
>>> import torch
>>> file_all = 'predict_predictions.pt'
>>> d_all = torch.load(file_all, map_location='cpu')
>>> d_all.keys()
dict_keys(['predictions', 'ids'])
>>> len(d_all['predictions']) # this equal to #sample
586
>>> len(d_all['ids']) # ids are numeric index corresponding to .h5ad file
586
>>> d_all['predictions'].shape # (B, L, D), L is max sequence length of all samples
torch.Size([586, 90, 128])
We can retrieve all the gene embedding and aggregate into cell embedding (like max pooling):
>>> d_all_maxpooling = [d_all['predictions'][i,:,:] for i in range(d_all['predictions'].shape[0])]
>>> d_all_maxpooling = [i[~torch.any(i.isnan(), dim=1)] for i in d_all_maxpooling]
>>> d_all_maxpooling = torch.cat([i.max(dim=0)[0].view(1,-1) for i in d_all_maxpooling])
>>> d_all_maxpooling.shape
torch.Size([586, 128])
Runs of genbio-ai GB.Tissue-3M on huggingface.co
20
Total runs
0
24-hour runs
1
3-day runs
1
7-day runs
-15
30-day runs
More Information About GB.Tissue-3M huggingface.co Model
Url of GB.Tissue-3M
GB.Tissue-3M huggingface.co Url
Provider of GB.Tissue-3M huggingface.co
Other API from genbio-ai
Total runs:
1.1K
Run Growth:
-1.1K
Growth Rate:
-97.46%
Total runs:
312
Run Growth:
-354
Growth Rate:
-113.46%
Total runs:
218
Run Growth:
50
Growth Rate:
22.94%
Total runs:
170
Run Growth:
101
Growth Rate:
60.12%
Total runs:
147
Run Growth:
-97
Growth Rate:
-65.99%
Total runs:
136
Run Growth:
52
Growth Rate:
38.24%
Total runs:
80
Run Growth:
-672
Growth Rate:
-819.51%
Total runs:
66
Run Growth:
-541
Growth Rate:
-819.70%
Total runs:
57
Run Growth:
-650
Growth Rate:
-1031.75%
Total runs:
55
Run Growth:
17
Growth Rate:
30.91%
Total runs:
30
Run Growth:
4
Growth Rate:
12.90%
Total runs:
29
Run Growth:
17
Growth Rate:
58.62%
Total runs:
29
Run Growth:
12
Growth Rate:
41.38%
Total runs:
24
Run Growth:
-8
Growth Rate:
-33.33%
Total runs:
22
Run Growth:
3
Growth Rate:
13.04%
Total runs:
18
Run Growth:
-13
Growth Rate:
-68.42%
Total runs:
15
Run Growth:
6
Growth Rate:
40.00%
Total runs:
15
Run Growth:
2
Growth Rate:
13.33%
Total runs:
14
Run Growth:
1
Growth Rate:
6.67%
Total runs:
13
Run Growth:
0
Growth Rate:
0.00%
Total runs:
13
Run Growth:
3
Growth Rate:
21.43%
Total runs:
13
Run Growth:
5
Growth Rate:
38.46%
Total runs:
12
Run Growth:
1
Growth Rate:
7.69%
Total runs:
12
Run Growth:
8
Growth Rate:
66.67%
Total runs:
12
Run Growth:
6
Growth Rate:
50.00%
Total runs:
12
Run Growth:
2
Growth Rate:
15.38%
Total runs:
11
Run Growth:
-6
Growth Rate:
-54.55%
Total runs:
11
Run Growth:
2
Growth Rate:
18.18%
Total runs:
11
Run Growth:
7
Growth Rate:
63.64%
Total runs:
11
Run Growth:
2
Growth Rate:
16.67%
Total runs:
10
Run Growth:
6
Growth Rate:
60.00%
Total runs:
10
Run Growth:
-10
Growth Rate:
-100.00%
Total runs:
10
Run Growth:
-4
Growth Rate:
-33.33%
Total runs:
9
Run Growth:
-5
Growth Rate:
-50.00%
Total runs:
9
Run Growth:
-17
Growth Rate:
-188.89%
Total runs:
9
Run Growth:
-7
Growth Rate:
-70.00%
Total runs:
9
Run Growth:
6
Growth Rate:
66.67%
Total runs:
9
Run Growth:
-1
Growth Rate:
-10.00%
Total runs:
9
Run Growth:
-8
Growth Rate:
-88.89%
Total runs:
8
Run Growth:
-5
Growth Rate:
-71.43%
Total runs:
8
Run Growth:
-266
Growth Rate:
-3325.00%
Total runs:
7
Run Growth:
5
Growth Rate:
62.50%
Total runs:
6
Run Growth:
-23
Growth Rate:
-383.33%
Total runs:
6
Run Growth:
4
Growth Rate:
66.67%
Total runs:
5
Run Growth:
4
Growth Rate:
80.00%
Total runs:
5
Run Growth:
3
Growth Rate:
60.00%
Total runs:
5
Run Growth:
5
Growth Rate:
100.00%
Total runs:
5
Run Growth:
1
Growth Rate:
20.00%
Total runs:
5
Run Growth:
0
Growth Rate:
0.00%
Total runs:
4
Run Growth:
-1
Growth Rate:
-25.00%
Total runs:
4
Run Growth:
-2
Growth Rate:
-50.00%
Total runs:
3
Run Growth:
-2
Growth Rate:
-66.67%
Total runs:
3
Run Growth:
-1
Growth Rate:
-25.00%
Total runs:
3
Run Growth:
-5
Growth Rate:
-166.67%
Total runs:
2
Run Growth:
-4
Growth Rate:
-200.00%
Total runs:
2
Run Growth:
-5
Growth Rate:
-250.00%
Total runs:
2
Run Growth:
-2
Growth Rate:
-100.00%
Total runs:
1
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%