Dear Dheeraj,

As you 'requested' a top 5 of best tools (besides the one you already used).

1. TOPCONS (my personal favourite)

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4489233/

and the description of the update:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4489233/

and a version to examine one single sequence:

https://academic.oup.com/bioinformatics/article/27/9/1322/242605

2. SCAMPI2: an improved version of SCAMPI

https://academic.oup.com/bioinformatics/article/32/8/1158/1743866

The authors have an interesting remark regarding TMHMM you used! I quote: “Finally, we found it interestingly that the method that is best at separating TM and non-TM proteins is the 15-years-old TMHMM, with 99% accuracy. However, the good separation comes at the price of non-optimal accurate topology predictions. The accuracy is 56 versus 63% for HMMTOP developed at the same time.”

3. MEMSAT3:

https://academic.oup.com/bioinformatics/article/23/5/538/239106

4. Heliquest combined with Heliquest based Eisenberg plot approach: If you want to use an “out of the box” approach you can use Heliquest. See the following reference for an example on how this works out:

https://www.researchgate.net/publication/230825800_Prediction_of_Lipid-Binding_Regions_in_Cytoplasmic_and_Extracellular_Loops_of_Membrane_Proteins_as_Exemplified_by_Protein_Translocation_Membrane_Proteins

And the corresponding supplement for more details:

https://www.researchgate.net/publication/328346727_SupplJMemBiol2013pdf

5. PRED-TMBB2: (suitable for beta-barrel outer membrane proteins)

https://academic.oup.com/bioinformatics/article/32/17/i665/2450774

Wishing you all the best with your research in this exciting field.