Compute_IC

Compute Information Content from a list of GO IDs

Install

• Create a new virtual environment python3 -m venv venv_name
• Activate new enviroment source venv_name/bin/activate
• Install requirements pip install -r requirements.txt

Input Format

#Accession\tGOid
lcl|CADEPI010000048.1_prot_CAB3369837.1_10820	GO:0000922
lcl|CADEPI010000048.1_prot_CAB3369837.1_10820	GO:0005876

Usage

• As a CLI app
  • Not providing GPAD annotations python compute_IC.py path/to/input/file
  • Providing GPAD annotations python compute_IC.py path/to/input/file --annotation path/to/gpad/file
• As a module
  • Simply import compute_IC and use the functions in your code

Outputs

With python compute_IC.py... it generates some output folders (outputs and plots)

• outputs/filename.tsv has the following format¹
  • Seq_ID: Sequence IDs as per input file entry
  • Category: biological_process, molecular_function or cellular_component
  • GOid: GO:XXXXXXX identifier
  • Descentant_count: Number of offpring for GO:XXXXXXX
  • Depth_level: Distance from root
  • Information content: Information content for GO:XXXXXXX
  • Description: description for GO:XXXXXXX

# Seq_ID\tCategory\tGOid\tDescendant_count\tDepth_level\tInformation_content\tDescription
lcl|CADEPI010000384.1_prot_CAB3384984.1_25967	molecular_function	GO:0005525	0	8	7.173966949456064	GTP binding
lcl|CADEPI010000015.1_prot_CAB3364279.1_5262	molecular_function	GO:0005509	4	5	6.926897219781138	calcium ion binding
lcl|CADEPI010000037.1_prot_CAB3368322.1_9305	molecular_function	GO:0003899	9	7	9.69997493502881	DNA-directed 5'-3' RNA polymerase activity
lcl|CADEPI010000066.1_prot_CAB3371947.1_12930	molecular_function	GO:0003697	8	5	9.39086641174421	single-stranded DNA binding
lcl|CADEPI010000037.1_prot_CAB3368322.1_9305	biological_process	GO:0006351	29	9	7.605963657147811	DNA-templated transcription
lcl|CADEPI010000066.1_prot_CAB3371947.1_12930	molecular_function	GO:0005524	0	8	5.127132583684966	ATP binding
lcl|CADEPI010000002.1_prot_CAB3359968.1_951	cellular_component	GO:0019773	2	2	11.524807227649527	proteasome core complex, alpha-subunit complex
lcl|CADEPI010000066.1_prot_CAB3371947.1_12930	biological_process	GO:0006281	46	7	6.919242348331503	DNA repair

• plots/Information_content_filename.jpeg represents Information Content (X axis) against probability density (Y axis) for BP, CC and MF.
  • Higher Information Content (right on the X axis) implies more specific GO term and a smaller (left on the X axis) value a more general GO term².
  • Information content is calculated using the formula²: $IC=-log{_2}{(p(t))}$
    • Where $p(t)= 1 - \frac{'Offspring\ Count'}{'Offspring\ Count'\ +\ 'Ancestors\ Count'}$

Footnotes

1. Klopfenstein, D.V., Zhang, L., Pedersen, B.S. et al. GOATOOLS: A Python library for Gene Ontology analyses. Sci Rep 8, 10872 (2018). https://doi.org/10.1038/s41598-018-28948-z ↩

2. Louie B, Bergen S, Higdon R, Kolker E. Quantifying protein function specificity in the gene ontology. Stand Genomic Sci. 2010 Mar 30;2(2):238-44. doi: 10.4056/sigs.561626. PMID: 21304708; PMCID: PMC3035283. ↩ ↩²