snap-stanford/SATURN

Issues

• One of the species is experiencing this issue; can not create XXX.pkl

  #69 opened a month ago by Joyce121
  1

• ValueError: cannot convert float NaN to integer in loss_string += [f"Avg Loss {species}: {round(np.average(epoch_ave_losses[species]))}"]

  #68 opened a month ago by micheladallangelo
  1

• Program running killed without errors

  #66 opened 2 months ago by yilevine
  2

• Out of memory error when using ESM2 for protein embeddings

  #56 opened 5 months ago by dariotommasini
  4

• What computer configuration is required to use this code

  #67 opened 2 months ago by idontwanttosayaword
  1

• trouble installing the package

  #63 opened 2 months ago by RanZhang08
  2

• question about the input sequences for esm

  #61 opened 2 months ago by huawen-poppy
  1

• Can I use spatial transcriptome data as input data?

  #65 opened 2 months ago by jekim2022
  1

• Can I integrate two mouse datasets together?

  #64 opened 2 months ago by vishnums007
  1

• Vertebrate-level brain atlas, questions about evolutionary distanct integration

  #62 opened 3 months ago by DiracZhu1998
  4

• Which seed is the best in your frog_zebrafish_embryogenesis analysis?

  #60 opened 4 months ago by vishnums007
  1

• What are seeds? What is the optimal number of seeds to run?

  #54 opened 5 months ago by vishnums007
  4

• The meaning of weights in macrogene?

  #59 opened 4 months ago by xuzhougeng
  1

• Example of label transfer with saturn?

  #58 opened 4 months ago by dariotommasini
  1

• SyntaxError while running saturn_multiple_seeds.py

  #55 opened 4 months ago by vishnums007
  1

• RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cpu)

  #57 opened 5 months ago by matejasoretic
  1

• How to preserve all the metadata information in the Anndata from two species to the Final AnnData after integration?

  #53 opened 5 months ago by vishnums007
  2

• Looking for "publicly available protein embedding datasets we have provided."

  #52 opened 5 months ago by vishnums007
  2

• Error when training saturn

  #51 opened 5 months ago by Xicici-Yan
  2

• Question: protein embedding, are we only using protein sequences?

  #50 opened 5 months ago by YaoLi3
  2

• .append() does not exist anymore in pandas

  #22 opened a year ago by anaellle
  6

• KeyError: 'species_gene_name'

  #49 opened 5 months ago by matejasoretic
  2

• train.SATURN ERROR. Is a GPU necessary?

  #48 opened 5 months ago by Tongbcc
  1

• New Data Integration

  #46 opened 6 months ago by cadyyuheng
  3

• issue with barcodes if species in the run.csv are provided in a non-alphabetic order

  #47 opened 6 months ago by matejasoretic
  1

• ValueError: b'Extrapolation not allowed with blending'

  #44 opened 6 months ago by cadyyuheng
  1

• unexpected keyword argument 'miner_type'

  #45 opened 6 months ago by matejasoretic
  0

• Should I must provide the score_adata ?

  #43 opened 6 months ago by bitcometz
  1

• Similar to KeyError: 'cell_type' #4

  #41 opened 6 months ago by matejasoretic
  3

• Failed for running frog_zebrafish_embryogenesis

  #42 opened 6 months ago by bitcometz
  1

• Within species integration

  #40 opened 6 months ago by cadyyuheng
  1

• Which ESM2 transformer should be used for proteome-wide embedding generation?

  #39 opened 6 months ago by matejasoretic
  1

• Please help with GSE113074_Corrected_combined.annotated_counts.h5ad

  #38 opened 6 months ago by bitcometz
  1

• Error with installation？

  #35 opened 6 months ago by bitcometz
  3

• Question about centroid creation

  #33 opened 6 months ago by otoky
  1

• Help with file "zebrafish_cell_types_mapping"

  #37 opened 6 months ago by bitcometz
  2

• Where is the extract.py file used in protein embedding?

  #27 opened 9 months ago by vinettey
  2

• Error at the biginning of the training step : centroid_weights.append(torch.tensor(species_genes_scores[sgn]))

  #36 opened 6 months ago by Hweeseon
  3

• Question about availability

  #34 opened 7 months ago by levinhein
  1

• creation of celltype_map csv

  #32 opened 7 months ago by otoky
  5

• train-saturn.py debugging

  #31 opened 7 months ago by otoky
  2

• Trouble creating protein embeddings

  #30 opened 7 months ago by otoky
  1

• Sapiens (~15gb) data not available

  #29 opened 7 months ago by dummyindex
  1

• Torch version not available

  #28 opened 9 months ago by dariotommasini
  1

• How indices_tuple is always empty tensors?

  #24 opened a year ago by MohammedZidane
  9

• Precomputed embeddings

  #19 opened a year ago by bschilder
  15

• Tunning Saturn

  #25 opened a year ago by Hrovatin
  3

• where in the code the weights from the genes layer to the macrogenes layer are frozen?

  #23 opened a year ago by MohammedZidane
  1

• Are the species inputs fed to the model together?

  #21 opened a year ago by MohammedZidane
  2

• For which purpose is use_batch_labels used for?

  #20 opened a year ago by cam1flag
  1