rohith-2 / chaos-game-representation_bioseq

Representation of Bio Sequences via Chaos Game and using the same to find similarities

Chaos-Game-Representation_BioSeq

Representation of Bio Sequences via Chaos Game and using the same to find similarities.

Medium : https://rrohith2001.medium.com/chaos-game-representation-of-genetic-sequences-e0e6bdcfaf6c

---

Data

Each excel file is a collection of all sequences in the respective family.

---

GUI

https://share.streamlit.io/rohith-2/chaos-game-representation_bioseq/stream.py

---

Gene-Similarity

CGR Matrix is a 2D matrix => (x,y) which consists of normalised value ranging from 0 to 1, which depicts the intensity of a color at any given (x,y)
The first two rows are considered for similarity measurement:

cgr_vec = Empty Vector()
for i <- cgr matrix of SEQ_1 # i iterates row wise
  a = max(i)
  new_row = i/a # Element-Wise Division
  cgr_vector = cgr_vector + new_row
  
cgr_vec_2 = Empty Vector()
for i <- cgr matrix of SEQ_2 # i iterates row wise
  a = max(i)
  new_row = i/a # Element-Wise Division
  cgr_vector_2 = cgr_vector_2 + new_row

Correlation(cgr_vec,cgr_vec_2)

cgr_vec and cgr_vec_2 will be vectors which can be utilised for measuring similarity via Spearmans correlation.

---

Running the GUI

The GUI is universally accesible via the above mentioned link to run in locally :

git clone https://github.com/Rohith-2/Chaos-Game-Representation_BioSeq.git
pip install -r requirements.txt  
cd Chaos-Game-Representation_BioSeq/GUI/
streamlit run gui_v2.py