PCCN

In this study, we proposed a protein co-conservation weighted network (PCCN) model simply based on protein sequence to investigate the mutations effects on the spike protein. We then compared the network topological features of mutation and non-mutation sites at the residue level and variant level. Finally, we mined the correlation of topological features with the mutation effects on spike protein stability changes and binding free energy changes.

Working Flow

1. Data Preparation.

   • S protein FASTA seqdata/YP_009724390.1.txt.
   • Variant of concern data/summary - 20211201.xlsx.

2. Find protein family .

   • Fasta seqs: data/protein-matching-IPR042578.fasta.
   • Detail information data/protein-matching-IPR042578.json .

3. Seqs of same protein family are too much, so filter seqs with same name and savefilter_fanmily_protein.py.

   • Fasta seqs: data/protein-matching-IPR042578.filter.fasta.
   • Detail information data/protein-matching-IPR042578.filter.csv.

4. clustalX2 MSA, and export as data/protein-matching-IPR042578.filter.fasta.aligned.

5. cal_procon.py calculates the conservation score, and result is data/procon/type{}.txt.

   • Due to the default output format is hard to analysis, it will be converted as csv filedata/procon/type{}_parse.csv.

6. analysis_procon.py finds the conservation of the variants, and save as data/procon/analysis.json.

7. v2/output_procon_analysis will construct the network and analysis the network.

Update log

---

2022.5.16

1. Backup the history result in data/v1/
2. Update the variants summary - 20220516
3. Run analysis_procon.py

---

2022.6.8.

Via reading the result of network, stability and affinity, find the relationship within them.

Network medium files:

• group distribution statistic information.xlsx: variants result
  • 1/4 1/2 3/4 quantile: quantile
  • mean: average value
  • score: variant score
  • t p: mean value of T-test and p
  • result
  • name: variant name
  • index
• aas_info.csv: network characteristics of postions

---

2022.7.9 ecdc doesn't provide complete mutations which need to be required

• https://www.ecdc.europa.eu/en/covid-19/variants-concern

Mutation source:

1. https://cov-lineages.org/ can query variant
2. https://outbreak.info/situation-reports?pango=B.1.617.2 can query mutation
   • url to get mutations: https://api.outbreak.info/genomics/lineage-mutations?pangolin_lineage=B.1.617.2&frequency=0.75

Flow

1. Update variant: https://www.ecdc.europa.eu/en/covid-19/variants-concern
2. Crawler mutation crawler.py
3. Filter data, remove:
   • duplication
   • without evidence

Network color

• Node
  • mutation #bf643b size 20
  • normal #008ea0 size 10
• Edge
  • neighbour #f66b0e
  • mutation #ffc300
  • normal #b7e5dd

---

2022.8.4.

1. Gephi
   • Only 2 color for edges
     • path and not path
     • ♥ mutation and no-mutation
   • position size: conservation score

---

...

---

2022.9.28

• Update network in Gephi
• Calculate DeepDGG for 6VXX and 6VYB
  • Try to submit part of mutation, but failed
  • So, calculate all possible mutations
    • 6VXX: http://protein.org.cn/jobstat.php?type=ddg&id=47029479
    • 6VYB: http://protein.org.cn/jobstat.php?type=ddg&id=87342993

---

2022.12.13

• Improve project