search

broadinstitute / chronos

Modeling of time series data for CRISPR KO experiments
BSD 3-Clause "New" or "Revised" License
27 stars 10 forks source link

received some nonnull values for genes in cell lines that have no guides targeting them, or inappropriate null values #17

Open Xinyi-Evelyn opened 6 days ago

Xinyi-Evelyn commented 6 days ago

serWarning: received some nonnull values for genes in cell lines that have no guides targeting them, or inappropriate null values warn( RuntimeWarning: invalid value encountered in subtract residue = pd.DataFrame(de.values - means).fillna(0).values verifying graph integrity verifying user inputs verifying variables

AssertionError Traceback (most recent call last) in Chronos.init(self, readcounts, guide_gene_map, sequence_map, negative_control_sgrnas, gene_effect_hierarchical, gene_effect_smoothing, kernel_width, gene_effect_L1, gene_effect_L2, offset_reg, excess_variance, guide_efficacy_reg, library_batch_reg, growth_rate_reg, smart_init, pretrained, cell_efficacy_guide_quantile, initial_screen_delay, scale_cost, max_learning_rate, dtype, verify_integrity, log_dir, to_normalize_readcounts) 958 if verify_integrity: 959 print("\tverifying graph integrity") --> 960 self.nan_check() 962 self.epoch = 0 964 if self._pretrained:

1827 print('verifying variables') 1828 if pd.isnull(self.sess.run(self._combined_gene_effect, self.run_dict)).sum().sum() > 0: -> 1829 assert False, "nulls found in self._combined_gene_effect" 1830 if pd.isnull(self.sess.run(self.v_guide_efficacy, self.run_dict)).sum().sum() > 0: 1831 assert False, "nulls found in self.v_guide_efficacy"

AssertionError: nulls found in self._combined_gene_effect

joshdempster commented 6 days ago

Hi Xinyi-Evelyn, can you share the full set of code you ran that produced this error?